Numerical study of particle-size distributions retrieved from angular light-scattering data using an evolution...

Area of Science:

• Physics
• Applied Mathematics
• Materials Science

Background:

• Particle size distribution is crucial for understanding material properties.
• Angular light-scattering data is a common method for particle characterization.
• Existing inversion methods often require a priori information, limiting their applicability.

Purpose of the Study:

• To develop and validate an evolution strategy algorithm for retrieving particle size distributions.
• To compare the algorithm's accuracy against established methods like Chin-Shifrin.
• To assess the algorithm's ability to avoid the need for prior distribution domain information.

Main Methods:

• Utilized Mie theory to generate theoretical angular light-scattering intensity patterns.
• Employed an evolution strategy algorithm to solve the inverse problem.
• Tested the algorithm using known normal, gamma, and lognormal distributions within a specific modal size parameter range (100 ≤ α ≤ 150).

Main Results:

• The evolution strategy algorithm successfully retrieved known particle size distributions.
• The algorithm demonstrated higher accuracy compared to the Chin-Shifrin inversion method.
• The proposed method effectively reconstructed distributions without requiring a priori knowledge of the size domain.

Conclusions:

• Evolution strategies are a viable and accurate approach for solving inverse problems in particle size distribution retrieval.
• This algorithm offers an advantage over traditional methods by eliminating the need for prior assumptions about the distribution.
• The findings support the broader application of evolutionary computation in optical particle characterization.