Optimal parameter design for power electronic converters using a probabilistic learning-based stochastic surrogate model

The selection of optimal design for power electronic converter parameters involves balancing efficiency and thermal constraints to ensure high performance without compromising safety. This study proposes a novel probabilistic-learning-based stochastic surrogate modeling framework, which simultaneously enables uncertainty-aware predictions and feasibility-guided optimization, a combination not commonly explored in prior converter design studies. The approach begins with a neural network classifier that evaluates the feasibility of parameter configurations, effectively filtering out unsafe and/or impractical inputs. Subsequently, a probabilistic prediction model estimates the converter’s efficiency and temperature while quantifying prediction uncertainty, providing both performance insights and reliability metrics. Finally, a heuristic optimization-based model is employed to optimize a multiobjective function that maximizes efficiency while adhering to thermal constraints. The optimization process incorporates penalty terms to discourage solutions that violate practical thresholds, ensuring actionable and realistic recommendations. An advanced heuristic optimization method is used to find the optimal solution and is compared with several well-known search algorithms, including genetic algorithm, particle swarm optimization, simulated annealing, tabu search, and stochastic hill climbing. The results demonstrate significant improvements in predictive accuracy and optimization outcomes, offering a robust solution for advancing power electronics design. The proposed framework is generalizable and can benefit the broader scientific community by serving as a scalable tool for fast, reliable design space exploration across diverse converter architectures and operating scenarios.