A Neural Network Based Approach to Simulate Electrothermal Device Interaction in SPICE Environment
| Autor: | Nicola Delmonte, Diego Chiozzi, Mirko Bernardoni, Paolo Cova |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Artificial neural network
Computer science 020208 electrical & electronic engineering Spice Semiconductor device modeling 02 engineering and technology Temperature measurement Nonlinear system Power electronics MOSFET 0202 electrical engineering electronic engineering information engineering Electronic engineering Feature (machine learning) Power semiconductor device Electrical and Electronic Engineering |
| Zdroj: | IEEE Transactions on Power Electronics. 34:4703-4710 |
| ISSN: | 1941-0107 0885-8993 |
| DOI: | 10.1109/tpel.2018.2863186 |
| Popis: | An innovative modeling methodology for the simulation of electrothermal interaction in power devices, based on neural networks (NNs), is shown. The suitability of NNs in modeling the complicated nonlinear, temperature dependent characteristic that power electronics devices feature is shown. The proposed methodology is particularly suited to be implemented in electrical simulators. The approach can be divided in two parallel steps: first, NNs are used to describe the complex, highly nonlinear electrothermal characteristic of the considered device; and second, a nonlinear RC -based thermal model is generated, with a method published in a previous work. These two subsystems are coupled together in order to achieve a self-consistent electrothermal model. The modeling results are validated against experiments with very satisfactory results. The technique is explained in detail; advantages and limitations of the method are then discussed. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|