Comprehensive Parameter Optimization Using an Empowered and Lightweight Surrogate Model

Fine tuning the parameters is crucial for achieving high-performance power electronics converters. Traditionally, iterative testing using professional simulation tools has been a common approach. However, running the simulation model is time consuming, and online parameter optimization generates par...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on power electronics 2024-10, Vol.39 (10), p.12124-12129
Main Authors: Yang, Qifan, Huang, Dihong, Chen, Yong, Dai, Ningyi
Format: Article
Language:English
Subjects:
Computational modeling
Control parameter tuning
Genetic algorithms
Optimization
Phase locked loops
Power electronics
Real-time systems
simulation
soft open point (SOP)
surrogate model
Tuning
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Fine tuning the parameters is crucial for achieving high-performance power electronics converters. Traditionally, iterative testing using professional simulation tools has been a common approach. However, running the simulation model is time consuming, and online parameter optimization generates parameters specific to each operating condition. In this article, we propose a novel approach that combines artificial intelligence (AI)-aided parameter tuning with simulation using a data-driven empowered surrogate model. The surrogate model is trained using a dataset derived from 3000 simulation tests, enabling rapid parameter tuning with feedback on system performance within a time frame of less than 0.1 ms, even on devices with restricted computational capabilities. Moreover, comprehensive parameter optimization for multiscenarios can be achieved using the surrogate model. A case study focusing on the parameter tuning of the soft-open-point is provided, including a comparison with AI-aided autonomous online parameter tuning methods. The results demonstrate the effectiveness and efficiency of the proposed approach.
ISSN:0885-8993
1941-0107
DOI:10.1109/TPEL.2024.3396504