Switching Converter Circuits Representation in CCM & DCM Using Graph Neural Network

• Published in: Conference proceedings - IEEE Applied Power Electronics Conference and Exposition pp. 2697 - 2702
• Main Authors: Khamis, Ahmed K., Agamy, Mohammed
• Format: Conference Proceeding
• Language: English
• Published: IEEE 19.03.2023
• Subjects: Bond Graph; Circuit synthesis; Electric circuit; Graph neural networks; Graph Neural Networks (GNN); Layout; Machine Learning; Switches; Switching converters; Topology; Training
• ISSN: 2470-6647
• DOI: 10.1109/APEC43580.2023.10131334

This paper proposes a method of transferring physical converter circuits working in continuous conduction mode (CCM) and discontinuous conduction mode (DCM) to graph representation, independent of the connection or the number of circuit components, so that machine learning (ML) algorithms and applications can be easily applied. Such methodology is applicable to converter circuits with any number of switches, components, sources and loads, and can be useful in applications such as artificial intelligence (AI) based circuit synthesis, automatic circuit layout generation, AI circuit design automation and many others. Three of the most common converters (Buck, Boost, and Buck-boost) are utilized as examples of how to apply the proposed methodology and as a proof of concept. These converters are mapped to ML space then a six class classifier (for three converter topologies in CCM and DCM) is applied that results in a training and testing accuracy of 100%.