Elmer FEM-Dakota: A unified open-source computational framework for electromagnetics and data analytics

TL;DR

This paper presents Elmer FEM-Dakota, an open-source integrated framework combining electromagnetics and data analytics, enabling advanced uncertainty quantification and model analysis for electric machine design.

Contribution

It introduces a novel open-source interface unifying electromagnetics and data analytics tools, validated against benchmarks, for improved design and analysis of electric machines.

Findings

Successful validation against benchmark tests

Enhanced understanding of model predictions

Facilitated time-sensitive electric machine design

Abstract

Open-source electromagnetic design software, Elmer FEM, was interfaced with data analytics toolkit, Dakota. Furthermore, the coupled software was validated against a benchmark test. The interface developed provides a unified open-source computational framework for electromagnetics and data analytics. Its key features include uncertainty quantification, surrogate modelling and parameter studies. This framework enables a richer understanding of model predictions to better design electric machines in a time sensitive manner.