Multi-Output Variable-Fidelity Bayesian Optimization of a Common Mode Choke

Jan  Hansen; Rodrigo Silva Rezende; Rolf Schuhmann; Mirsad Hadziefendic

doi:10.1109/EMC/SI/PI/EMCEurope52599.2021.9559196

Multi-Output Variable-Fidelity Bayesian Optimization of a Common Mode Choke

Jan Hansen
, Rodrigo Silva Rezende^*
, Rolf Schuhmann
, Mirsad Hadziefendic

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review

Abstract

In this paper, we propose a modification of the well-known kriging modeling technique and its multi-fidelity variant in order to generate a surrogate model of a vector-valued output of a common mode choke. Combined with Bayesian optimization, we solve a multi-objective optimization problem to find the tradeoff between inductance, total volume, and resonance frequency. With the proposed method, we calculate a Pareto front of the objective functions more than 200 times faster compared to the solution without the surrogate.

Original language	English
Title of host publication	2021 IEEE International Joint EMC/SI/PI and EMC Europe Symposium
Pages	1093-1098
ISBN (Electronic)	978-1-6654-4888-8
DOIs	https://doi.org/10.1109/EMC/SI/PI/EMCEurope52599.2021.9559196
Publication status	Published - 2 Sept 2021
Externally published	Yes
Event	2021 Joint IEEE International Symposium on Electromagnetic Compatibility Signal and Power Integrity, and EMC Europe: EMC/SI/PI/EMC Europe 2021 - Raleigh, United States Duration: 26 Jul 2021 → 20 Aug 2021

Conference

Conference	2021 Joint IEEE International Symposium on Electromagnetic Compatibility Signal and Power Integrity, and EMC Europe
Abbreviated title	EMC/SI/PI/EMC Europe 2021
Country/Territory	United States
City	Raleigh
Period	26/07/21 → 20/08/21

Access to Document

10.1109/EMC/SI/PI/EMCEurope52599.2021.9559196

Cite this

Hansen, J, Silva Rezende, R, Schuhmann, R & Hadziefendic, M 2021, Multi-Output Variable-Fidelity Bayesian Optimization of a Common Mode Choke. in 2021 IEEE International Joint EMC/SI/PI and EMC Europe Symposium. pp. 1093-1098, 2021 Joint IEEE International Symposium on Electromagnetic Compatibility Signal and Power Integrity, and EMC Europe, Raleigh, United States, 26/07/21. https://doi.org/10.1109/EMC/SI/PI/EMCEurope52599.2021.9559196

@inproceedings{1993dffcff8d4f98861a2fada32c3d15,

title = "Multi-Output Variable-Fidelity Bayesian Optimization of a Common Mode Choke",

abstract = "In this paper, we propose a modification of the well-known kriging modeling technique and its multi-fidelity variant in order to generate a surrogate model of a vector-valued output of a common mode choke. Combined with Bayesian optimization, we solve a multi-objective optimization problem to find the tradeoff between inductance, total volume, and resonance frequency. With the proposed method, we calculate a Pareto front of the objective functions more than 200 times faster compared to the solution without the surrogate.",

author = "Jan Hansen and \{Silva Rezende\}, Rodrigo and Rolf Schuhmann and Mirsad Hadziefendic",

year = "2021",

month = sep,

day = "2",

doi = "10.1109/EMC/SI/PI/EMCEurope52599.2021.9559196",

language = "English",

pages = "1093--1098",

booktitle = "2021 IEEE International Joint EMC/SI/PI and EMC Europe Symposium",

note = "2021 Joint IEEE International Symposium on Electromagnetic Compatibility Signal and Power Integrity, and EMC Europe : EMC/SI/PI/EMC Europe 2021, EMC/SI/PI/EMC Europe 2021 ; Conference date: 26-07-2021 Through 20-08-2021",

}

TY - GEN

T1 - Multi-Output Variable-Fidelity Bayesian Optimization of a Common Mode Choke

AU - Hansen, Jan

AU - Silva Rezende, Rodrigo

AU - Schuhmann, Rolf

AU - Hadziefendic, Mirsad

PY - 2021/9/2

Y1 - 2021/9/2

N2 - In this paper, we propose a modification of the well-known kriging modeling technique and its multi-fidelity variant in order to generate a surrogate model of a vector-valued output of a common mode choke. Combined with Bayesian optimization, we solve a multi-objective optimization problem to find the tradeoff between inductance, total volume, and resonance frequency. With the proposed method, we calculate a Pareto front of the objective functions more than 200 times faster compared to the solution without the surrogate.

AB - In this paper, we propose a modification of the well-known kriging modeling technique and its multi-fidelity variant in order to generate a surrogate model of a vector-valued output of a common mode choke. Combined with Bayesian optimization, we solve a multi-objective optimization problem to find the tradeoff between inductance, total volume, and resonance frequency. With the proposed method, we calculate a Pareto front of the objective functions more than 200 times faster compared to the solution without the surrogate.

U2 - 10.1109/EMC/SI/PI/EMCEurope52599.2021.9559196

DO - 10.1109/EMC/SI/PI/EMCEurope52599.2021.9559196

M3 - Conference paper

SP - 1093

EP - 1098

BT - 2021 IEEE International Joint EMC/SI/PI and EMC Europe Symposium

T2 - 2021 Joint IEEE International Symposium on Electromagnetic Compatibility Signal and Power Integrity, and EMC Europe

Y2 - 26 July 2021 through 20 August 2021

ER -

Multi-Output Variable-Fidelity Bayesian Optimization of a Common Mode Choke

Abstract

Conference

Access to Document

Fingerprint

Cite this