Skip to main content
APEC 2025 Main banner
Back

Technical Dialogue

Resonant Converters

D04.2 - cGANET-Enhanced Voltage Gain Modeling: Elevating CLLC Converter Accuracy

Thursday, March 20, 2025
11:30 AM – 1:30 PM ET
Location: Level One, Hall A3
The voltage gain model of resonant converters is vital for design and optimization. Traditional methods like Fundamental Harmonic Approximation (FHA) lack accuracy, and Time-Domain Analysis (TDA) is computationally intensive and varies across operating modes. This paper introduces a voltage gain prediction model using conditional Generative Adversarial Networks (cGANET). The model is trained with a dataset from simulations under various conditions, allowing it to accurately predict complex, non-linear interactions within the resonant converter without relying on simplifying assumptions or mode-specific formulas like FHA and TDA. Thanks to its strong extrapolation and interpolation abilities, the cGANET model achieves high accuracy within and beyond the training range, adapting to new conditions without extensive reconfiguration. Experimental validation with a prototype CLLC converter across various switching frequencies demonstrated a 60% improvement in accuracy under the worst conditions compared to FHA, proving its value for enhancing resonant converter design and optimization.