Modified FEA and ExtraTree algorithm for transformer Green's function modelling

TL;DR

This paper proposes a hybrid method combining modified finite element analysis and an ExtraTree machine learning algorithm to efficiently estimate a transformer's Green's function, reducing computation time while maintaining accuracy.

Contribution

It introduces a novel approach integrating FEA with ExtraTree and genetic algorithms to speed up Green's function estimation for transformers.

Findings

Reduced FEA computation time significantly

Maintained estimation accuracy with the hybrid method

Effective prediction of frequency response functions

Abstract

The Green's function of a transformer is essential for prediction of its vibration. As the Green's function cannot be measured directly and completely, the finite element analysis (FEA) is typically used for its estimation. However, because of the complexity of the transformer structure, the calculations involved in FEA are time consuming. Therefore, in this paper, a method based on FEA modified by an extremely random tree algorithm call ExtraTree is proposed to efficiently estimate the Green's function of a transformer. A subset of the frequency response functions from FEA will be selected by a genetic algorithm that can well present the structural variation. The FEA calculation time can be reduced by simply calculating the frequency response functions on this subset and predicting remainder using the trained ExtraTree model. The errors introduced in this method can be estimated from the corresponding frequency and the genetic algorithm error.