*WINNER* AI Based Optimization of Solid State Transformer Core for Modern Electric Vehicles Using Multi-Objective Genetic Algorithm

Abstract

Solid-state transformers are increasingly becoming a desirable alternative to traditional low frequency transformers due to their compact size and high efficiency, particularly in the field of Electric Vehicles (EV), which has seen rapid growth in recent years. This research offers a multi-objective AI-based high-frequency transformer (HFT) design optimization for solid-state transformer (SST) applications. As the key component of the SST, the optimization of the HFT design parameters is crucial for achieving high efficiency and power density, independent of its topology. The HFT is designed using a multi-objective Non-dominated Sorting optimization technique that reduces core volume (maximizing power density), total transformer losses, and overall cost from the set multiple Pareto-optimal solutions (POS). An 750kHz, 10kW HFT of different high permeability core materials is explored as a case study and the POS are presented. The findings show how the various design variables affect the goal functions. The results further show that the size, efficiency, and cost of the HFT may be efficiently optimized by carefully selecting design variables using the suggested method. A large number of the Pareto-Optimal solutions demonstrate that in the HFT design for SST applications, an efficiency of above 97% can be attained.