A methodology for the optimized design of power transformer insulation system

Purpose The purpose of this paper is to present a methodology based on an optimizer linked with electric finite element method (FEM) for automating the optimized design of power transformer insulation system structures. Design/methodology/approach The proposed methodology combines two stages to obta...

Full description

Saved in:
Bibliographic Details
Published in:Compel 2018-05, Vol.37 (3), p.1002-1010
Main Authors: Rodriguez-Ignacio, Hugo, Lopez-Fernandez, Xose M, Álvarez-Mariño, Casimiro
Format: Article
Language:English
Subjects:
Automation
Design engineering
Design optimization
Electric fields
Finite element method
Insulation
Methodology
Repair & maintenance
Safety margins
Transformers
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Purpose The purpose of this paper is to present a methodology based on an optimizer linked with electric finite element method (FEM) for automating the optimized design of power transformer insulation system structures. Design/methodology/approach The proposed methodology combines two stages to obtain the optimized design of transformer insulation system structures. First, an analytical calculation is carried out with the optimizer to search a candidate solution. Then, the candidate solution is numerically checked in detail to validate its consistency. Otherwise, these two steps are iteratively repeated until the optimizer finds a candidate solution according to the objective function. Findings The solutions found applying the proposed methodology reduce the inter-electrode distances compared to those insulation designs referenced in the literature for the same value of safety margin. Originality/value The proposed methodology explores a wide range of insulation system structures in an automated way which is not possible to do with the classical trial-and-error approach based on personal expertise.
ISSN:0332-1649
2054-5606
DOI:10.1108/COMPEL-12-2016-0561