Optimization of a SiC MOSFET behavioural circuit model by using a multi-objective genetic algorithm
| Autor: | Gaetano Bazzano, Giovanni Susinni, Alessandra Raffa, Nunzio Salerno, Santi Agatino Rizzo, PierPaolo Veneziano |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Optimal design
Computer science 05 social sciences Automotive 020207 software engineering Control engineering Context (language use) power device 02 engineering and technology stochastic optimization Power (physics) Spice Tree (data structure) power conversion Genetic algorithm 0202 electrical engineering electronic engineering information engineering SiC MOSFET Inverter widebandgap 0501 psychology and cognitive sciences Stochastic optimization Power semiconductor device 050107 human factors |
| Zdroj: | 2020 IEEE Energy Conversion Congress and Exposition (ECCE). |
| DOI: | 10.1109/ecce44975.2020.9235767 |
| Popis: | The use of proper models of the power devices is important when the design of the power conversions systems is performed. The optimal design requires high accuracy of the power device behaviour and low computational effort, especially fast simulation. Obtain quickly the power device model endowed with the previous two features is an important additional target of manufactures. Finally, they also want provide a tool which has to be effortlessly handled by the users. In this context, the behavioural model is useful since it includes a set of equations that emulates the device behaviour at its terminals and it is easy to implement in simulation tools (e.g. SPICE). Thus tree targets are reached: best compromise between accuracy and simulation burden, friendly-user. The major issue is to choose quickly the model parameters. In this work, this issue is taken on by means of a multi-objective genetic algorithm. A double test has been used to compare the measured and simulated waveforms. Such a comparison in an inverter module has highlighted the accurateness of the proposed solution. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|