Add elastic.py with Boussinesq point load solution for photoelastic simulations

[Copilot]

Implements Boussinesq point load solution for elastic half-space, parallel to existing Brazil test in disk.py. Generates synthetic stress fields and RGB polarized images for elastic contact problems.

Implementation

• photoelastimetry/elastic.py: Core module with analytical Boussinesq solution

  • boussinesq_stress_cartesian(): Computes σ_xx, σ_yy, τ_xy for point load P on semi-infinite half-space
  • generate_synthetic_boussinesq(): Produces synthetic photoelastic images via Mueller matrix formalism
  • post_process_synthetic_data(): 16-panel visualization (stress components, fringes, polarimetry)
  • Enforces boundary condition (zero stress above surface y < 0)

• json/elastic.json5: Configuration with material properties (C, thickness, wavelengths, Poisson's ratio)

• tests/test_elastic.py: 9 tests covering stress computation, boundary conditions, synthetic image generation, parameter validation

Usage

from photoelastimetry.elastic import boussinesq_stress_cartesian

# Compute stress field for 100N point load
sigma_xx, sigma_yy, tau_xy = boussinesq_stress_cartesian(X, Y, P=100.0, nu_poisson=0.3)

Or run as script to generate TIFF outputs:

python -m photoelastimetry.elastic
# Creates images/elastic/boussinesq_synthetic_{stress,images}.tiff

Technical notes

• Uses classical Boussinesq elasticity solution with singularity handling at load point
• Consistent with disk.py architecture: analytical solution → stress field → Mueller matrices → polarized intensities
• Supports configurable Poisson's ratio and multi-wavelength RGB imaging

Original prompt

add a new elastic.py file that will operate similarly to disk
pt, but for elastic solutions, let's start with the boussinesq point load on an elastic half space. when you run the file it should generate tiff files with synthetic images of the stress field and RGB polarised images.

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[Copilot]

Pull request overview

This PR adds a new elastic.py module implementing the Boussinesq point load solution for elastic half-space photoelastic simulations, parallel to the existing Brazil test in disk.py. The implementation includes analytical stress field computation, synthetic image generation via Mueller matrix formalism, and comprehensive visualization tools.

Changes:

• New analytical elastic solution module with Boussinesq stress computation
• Configuration file for elastic simulation parameters
• Comprehensive test suite with 9 test cases covering stress computation, boundary conditions, and parameter validation

Reviewed changes

Copilot reviewed 3 out of 4 changed files in this pull request and generated 4 comments.

File	Description
photoelastimetry/elastic.py	Core module implementing Boussinesq analytical solution, synthetic image generation, and 16-panel visualization
json/elastic.json5	Configuration file with material properties and simulation parameters
tests/test_elastic.py	Test suite covering stress field computation, boundary conditions, and synthetic data generation
.gitignore	Adds patterns for generated output files and images directory

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The variable name wavelengths_nm suggests nanometers, but the values are in meters (650e-9, 550e-9, 450e-9) and the comment confirms "in meters". This is inconsistent with the equivalent fixture in test_disk.py line 23 which correctly uses nanometer values (650, 550, 450) with the same variable name. The variable should either be renamed to wavelengths_m or the values should be changed to [650, 550, 450] to match the naming convention and be consistent with test_disk.py.

[Copilot]

The stress-optic coefficient values [3e-9, 3e-9, 3e-9] are identical for all three wavelengths (R, G, B), which differs from the pattern in test fixtures and other JSON files that use wavelength-dependent values like [2e-12, 2.2e-12, 2.5e-12] in test_disk.py or json/test.json5. While this may be intentional, using identical C values removes the wavelength-dependent behavior that is typically expected in photoelastic simulations. Consider using wavelength-dependent values like [3e-9, 3.3e-9, 3.75e-9] to better represent realistic material dispersion, or add a comment explaining why identical values are appropriate for this elastic simulation.

Suggested change

	C : [3e-9, 3e-9, 3e-9], // stress-optic coefficient for each wavelength (1/Pa)
	C : [3e-9, 3.3e-9, 3.75e-9], // stress-optic coefficient for each wavelength (1/Pa)

[Copilot]

The comment says "wavelengths, 4 polarizer angles" but according to line 170 the hardcoded shape is (height, width, 3, 4) in disk.py, while here it dynamically uses n_wavelengths. The comment should be more explicit about the array dimensions, like "# (height, width, n_wavelengths, 4 polarizer angles)" to match the actual shape tuple for clarity.

Suggested change

	synthetic_images = np.empty((height, width, n_wavelengths, 4)) # wavelengths, 4 polarizer angles
	synthetic_images = np.empty((height, width, n_wavelengths, 4)) # (height, width, n_wavelengths, 4 polarizer angles)

[Copilot]

The array is initialized with np.empty() which leaves uninitialized memory. While this is immediately populated in the loop below, using np.zeros() would be safer and make debugging easier if the loop fails to populate all values. This is consistent with defensive programming practices, especially for arrays that will be saved to files.

Suggested change

	synthetic_images = np.empty((height, width, n_wavelengths, 4)) # wavelengths, 4 polarizer angles
	synthetic_images = np.zeros((height, width, n_wavelengths, 4)) # wavelengths, 4 polarizer angles