Overview

1. Model

The Model introduces the implemented Population Balance Equation (PBE) solvers, including:

  • DPBESolver (grid-based, deterministic discretization)

  • PBMSolver (moment-based methods: QMOM, GQMOM, CQMOM)

  • MCPBESolver (Monte Carlo stochastic simulation)

  • ExtruderPBESolver (multi-region extension of DPBE)

It also explains the physical basis of PBE, the algorithms behind each solver, and provides a performance and stability comparison across methods.

2. Class Structure

The Class Structure documentation describes:

  • The design principles of all solver classes

  • Their modular organization (core, postprocessing, visualization)

  • Key attributes and methods essential for extending or customizing the solvers

This section helps users understand the internal logic of the library and how the main solvers are constructed.

3. Config Data

The Config Data section explains:

  • The role of configuration files in setting solver parameters

  • Why configs are used (to ensure reproducibility and flexible parameter control)

  • How to override defaults using __init__, config files, or manual attribute changes

  • Priority rules among these methods

This provides a unified way to manage simulation parameters consistently across different solvers.

4. Optimization Framework

The Optimization Framework documentation introduces:

  • The structure of the optimization modules (kernel_opt, kernel_opt_extruder)

  • How Ray Tune is used to enable scalable, parallel optimization

  • Methods to integrate solver runs into optimization loops

  • Guidelines on defining objective functions and interpreting optimization results

This section shows how to couple the PBE solvers with large-scale parameter studies and optimization tasks.