Solid-State Transformer

AI-driven design of a modular SST — from cell-level converter optimisation to system-level integration and control.

Read the complete three-stage SST learning series →

The linked articles are analytical teaching drafts. They state their assumptions and distinguish planned models and hardware from completed validation.

Why Solid-State Transformers?

Solid-state transformers replace bulky line-frequency transformers with compact, controllable power electronics — enabling bidirectional power flow, voltage regulation, and seamless integration of renewables, storage, and DC loads.

Multi-stage architecture

AC-DC rectification, isolated DC-DC conversion (DAB), and DC-AC inversion — each stage co-optimised by the AI Agent for the overall SST mission profile.

Modular cell design

Cascaded H-bridge or modular multi-level cells with AI-driven device selection, magnetics sizing, and thermal balancing across all modules.

Medium-voltage operation

For medium-voltage designs, insulation coordination, dv/dt management and module voltage sharing require dedicated engineering and validation.

AI Agent in Action

Topology exploration

Automated screening of DAB, LLC, and resonant CLLC cells — evaluating efficiency, power density, and fault tolerance trade-offs for the isolation stage.

Control co-design

Hierarchical control synthesis: cell-level soft-switching and current balancing, stage-level voltage regulation, and system-level power flow management.

Validation & feedback

Auto-generated test plans for each SST module — efficiency mapping, thermal cycling, and fault injection — feeding results back to refine the AI Agent's models.