A Computationally Efficient and Stable Learning-Based Controller for DC/AC Inverter
A new machine-learning-based control strategy for the dc/ac inverter is proposed in this article, which is highly computationally efficient and insensitive to model parameter variations. It provides fast offline neural network (NN) training and low computational cost for online digital signal processor (DSP)-based implementation. The recurrent equilibrium network (REN) is employed to achieve excellent transient and steady-state performance and the closed-loop system is proven to be asymptotically stable. The performance of the proposed approach is verified through experimental comparisons with the existing control methods.
Authors
- Herbert Ho‐Ching Iu (ORCID: https://orcid.org/0000-0002-0687-4038)
- Joshua Watts (ORCID: https://orcid.org/0009-0002-0494-2663)
- Tianhao Qie (ORCID: https://orcid.org/0000-0002-4463-5927)
- Tyrone Fernando (ORCID: https://orcid.org/0000-0003-0140-8887)
- Yun Liu (ORCID: https://orcid.org/0000-0002-1970-1291)
- Xinan Zhang (ORCID: https://orcid.org/0000-0002-9472-8785)
- Ran Li (ORCID: https://orcid.org/0000-0002-5813-5755)
- Ruigang Wang (ORCID: https://orcid.org/0000-0002-9954-4547)
- Wendong Feng (ORCID: https://orcid.org/0009-0009-2609-1311)
Institutions
- The University of Sydney (AU)
- Queensland University of Technology (AU)
- The University of Western Australia (AU)
Publication Details
- Journal
- IEEE Journal of Emerging and Selected Topics in Power Electronics
- Published
- 2025-02-07
- DOI
- https://doi.org/10.1109/jestpe.2025.3539984
- Primary Topic
- Sensorless Control of Electric Motors
- Type
- article
- Field-Weighted Citation Impact
- 0.00