Abstract
This paper analyzes the potential of Artificial Neural Networks (ANNs) for the modeling and optimization of magnetic components and, specifically, inductors. After reviewing the basic properties of ANNs, several potential modeling and design workflows are presented. A hybrid method, which combines the accuracy of 3D Finite Element Method (FEM) and the low computational cost of ANNs, is selected and implemented. All relevant effects are considered (3D magnetic and thermal field patterns, detailed core loss data, winding proximity losses, coupled loss-thermal model, etc.) and the implemented model is extremely versatile (30 input and 40 output variables). The proposed ANN-based model can compute 50'000 designs per second with less than 3% deviation with respect to 3D FEM simulations. Finally, the inductor of a 2 kW DC-DC buck converter is optimized with the ANN-based workflow. From the Pareto fronts, a design is selected, measured, and successfully compared with the results obtained with the ANNs. The implementation (source code and data) of the proposed workflow is available under an open-source license.
Highlights
The past few years have seen a rapid development of artificial intelligence applications in research, engineering, and industry [1]
5) With the obtained sensitivity, the weights are updated in order to diminish the error. This update is controlled by GUILLOD ET AL.: ARTIFICIAL NEURAL NETWORK (ANN) BASED FAST AND ACCURATE INDUCTOR MODELING AND DESIGN TABLE 1 Inductor Scaling Laws: Artificial Neural Networks (ANNs) Performance
SELECTED ANN-BASED MODEL From the aforementioned methods and the goals defined in the introduction, an ANN-based inductor model, which features the same accuracy as 3D Finite Element Method (FEM) simulations with a massively reduced computational cost, is presented in detail in the following
Summary
The past few years have seen a rapid development of artificial intelligence applications in research, engineering, and industry [1]. The usage of convolution layers allows the usage of multi-dimensional data (matrix of tensor) and the detection of complex patterns Such ANNs are very well r suited to large problems with unstructured data. ANNs, so far, have been mainly used for fault diagnosis [2], [3] and control strategies [4], [5] Another important field of power electronics, requiring complex models and heavy computations, is the modeling and multi-objective optimization of converters and components (e.g., with respect to volume, mass, cost, efficiency) [10]–[13]. The flexibility of ANNs. Inductors are selected because magnetic components typically represent the bottleneck of multi-objective optimization (e.g., model complexity, computational cost, size and diversity of the design and performance spaces) [11]–[13], [18].
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
