SBO_HB_PFM_LLC_WOR

This code provides a Python-based simulation and optimization framework for the analysis and design of half-bridge LLC resonant converters operating over a wide output-voltage and load range. The framework uses PySpice, a Python interface to the ngspice circuit simulator, to perform cycle-by-cycle transient simulations of the LLC tank network, with automatic detection of steady-state convergence.

The code includes five demonstration examples: (1) a basic single transient simulation to verify the environment; (2) block-by-block simulation until steady-state convergence, with waveform and metrics plots; (3) peak-gain curve mapping (maximum achievable output voltage as a function of switching frequency) using Brent's root-finding method; (4) closed-loop frequency search for a target output voltage at a single operating point, with full convergence visualization; and (5) batch evaluation of multiple operating points in parallel, producing per-point waveform PDFs and a consolidated metrics CSV.

A genetic algorithm (GA) optimization module based on NSGA-II (via pymoo) is also included. It minimizes the resonant current RMS while enforcing zero-voltage switching (ZVS) and regulation margin constraints across a set of user-defined operating points, and exports the Pareto-front results for post-processing and visualization.

The framework is intended to support the design methodology described in the associated article and to allow reproduction of the simulation results presented therein.