Bayesian Experimental Design for LEDs using Gaussian Processes

Peter Forster; Sebastian Schops; Wil Schilders; Stephan Bockhorst; Maximilian Mevenkamp

doi:10.1109/THERMINIC60375.2023.10325903

Bayesian Experimental Design for LEDs using Gaussian Processes

Peter Forster, Sebastian Schops, Wil Schilders, Stephan Bockhorst, Maximilian Mevenkamp

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

Abstract

We present a novel design of experiments approach for LEDs based on Gaussian processes. The method aims to decrease the measurement effort associated with characterizing LEDs based on their spectral power distribution (SPD) and derived quantities of interest (QOIs), such as the luminous flux or color coordinates. It is both easy to implement and based on open source software. We showcase the approach on an example taken from automotive applications. For the considered example, we are able to decrease the total number of measurements by over 75%, while the model is able to predict the SPD and derived QOIs with relative errors of less than 5%.

Original language	English
Title of host publication	2023 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023
Publisher	Institute of Electrical and Electronics Engineers
ISBN (Electronic)	9798350318623
DOIs	https://doi.org/10.1109/THERMINIC60375.2023.10325903
Publication status	Published - 2023
Event	29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023 - Budapest, Hungary Duration: 27 Sept 2023 → 29 Sept 2023

Conference

Conference	29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023
Country/Territory	Hungary
City	Budapest
Period	27/09/23 → 29/09/23

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

experimental design
Gaussian processes
LEDs
spectral power distribution

Access to Document

10.1109/THERMINIC60375.2023.10325903

Cite this

@inproceedings{bad72fd69c134009959a98d66108044b,

title = "Bayesian Experimental Design for LEDs using Gaussian Processes",

abstract = "We present a novel design of experiments approach for LEDs based on Gaussian processes. The method aims to decrease the measurement effort associated with characterizing LEDs based on their spectral power distribution (SPD) and derived quantities of interest (QOIs), such as the luminous flux or color coordinates. It is both easy to implement and based on open source software. We showcase the approach on an example taken from automotive applications. For the considered example, we are able to decrease the total number of measurements by over 75%, while the model is able to predict the SPD and derived QOIs with relative errors of less than 5%.",

keywords = "experimental design, Gaussian processes, LEDs, spectral power distribution",

author = "Peter Forster and Sebastian Schops and Wil Schilders and Stephan Bockhorst and Maximilian Mevenkamp",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023 ; Conference date: 27-09-2023 Through 29-09-2023",

year = "2023",

doi = "10.1109/THERMINIC60375.2023.10325903",

language = "English",

booktitle = "2023 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

}

Forster, P, Schops, S, Schilders, W, Bockhorst, S & Mevenkamp, M 2023, Bayesian Experimental Design for LEDs using Gaussian Processes. in 2023 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023. Institute of Electrical and Electronics Engineers, 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023, Budapest, Hungary, 27/09/23. https://doi.org/10.1109/THERMINIC60375.2023.10325903

Bayesian Experimental Design for LEDs using Gaussian Processes. / Forster, Peter; Schops, Sebastian; Schilders, Wil et al.
2023 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023. Institute of Electrical and Electronics Engineers, 2023.

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

TY - GEN

T1 - Bayesian Experimental Design for LEDs using Gaussian Processes

AU - Forster, Peter

AU - Schops, Sebastian

AU - Schilders, Wil

AU - Bockhorst, Stephan

AU - Mevenkamp, Maximilian

PY - 2023

Y1 - 2023

N2 - We present a novel design of experiments approach for LEDs based on Gaussian processes. The method aims to decrease the measurement effort associated with characterizing LEDs based on their spectral power distribution (SPD) and derived quantities of interest (QOIs), such as the luminous flux or color coordinates. It is both easy to implement and based on open source software. We showcase the approach on an example taken from automotive applications. For the considered example, we are able to decrease the total number of measurements by over 75%, while the model is able to predict the SPD and derived QOIs with relative errors of less than 5%.

AB - We present a novel design of experiments approach for LEDs based on Gaussian processes. The method aims to decrease the measurement effort associated with characterizing LEDs based on their spectral power distribution (SPD) and derived quantities of interest (QOIs), such as the luminous flux or color coordinates. It is both easy to implement and based on open source software. We showcase the approach on an example taken from automotive applications. For the considered example, we are able to decrease the total number of measurements by over 75%, while the model is able to predict the SPD and derived QOIs with relative errors of less than 5%.

KW - experimental design

KW - Gaussian processes

KW - LEDs

KW - spectral power distribution

UR - http://www.scopus.com/inward/record.url?scp=85179624820&partnerID=8YFLogxK

U2 - 10.1109/THERMINIC60375.2023.10325903

DO - 10.1109/THERMINIC60375.2023.10325903

M3 - Conference contribution

AN - SCOPUS:85179624820

BT - 2023 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023

PB - Institute of Electrical and Electronics Engineers

T2 - 29th International Workshop on Thermal Investigations of ICs and Systems, THERMINIC 2023

Y2 - 27 September 2023 through 29 September 2023

ER -

Bayesian Experimental Design for LEDs using Gaussian Processes

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this