L-Bot: a physically motivated deep learning based inductor modeling tool

Uttung Surange; Sourabh Khandelwal

doi:10.1109/APMC52720.2021.9661689

L-Bot: a physically motivated deep learning based inductor modeling tool

Uttung Surange, Sourabh Khandelwal

School of Engineering

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

1 Citation (Scopus)

Abstract

In this paper, we present L-BOT, a physically motivated automatic inductor modeling tool based on deep learning algorithm. This tool uses the physical topology of inductor models representing the winding resistance, core losses, and capacitive effects in the inductor. The parameters of physical topology are automatically extracted using L-BOT after training the deep learning based engine. We show excellent accuracy of the inductor models generated from L-BOT for six different commercial radio-frequency (RF) inductors.

Original language	English
Title of host publication	2021 IEEE Asia-Pacific Microwave Conference (APMC)
Place of Publication	Piscataway, NJ
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	82-84
Number of pages	3
ISBN (Electronic)	9781665437820
ISBN (Print)	9781665437837
DOIs	https://doi.org/10.1109/APMC52720.2021.9661689
Publication status	Published - 2021
Event	2021 IEEE Asia-Pacific Microwave Conference, APMC 2021 - Virtual, Australia Duration: 28 Nov 2021 → 1 Dec 2021

Conference

Conference	2021 IEEE Asia-Pacific Microwave Conference, APMC 2021
Country/Territory	Australia
City	Virtual
Period	28/11/21 → 1/12/21

Keywords

Machine Learning
Deep Learning
Inductor
Coilcraft
Neural Network
LSTM
s-parameters
parameter prediction

Access to Document

10.1109/APMC52720.2021.9661689

Cite this

@inproceedings{d526f927f22447b994294c8d2d647ddd,

title = "L-Bot: a physically motivated deep learning based inductor modeling tool",

abstract = "In this paper, we present L-BOT, a physically motivated automatic inductor modeling tool based on deep learning algorithm. This tool uses the physical topology of inductor models representing the winding resistance, core losses, and capacitive effects in the inductor. The parameters of physical topology are automatically extracted using L-BOT after training the deep learning based engine. We show excellent accuracy of the inductor models generated from L-BOT for six different commercial radio-frequency (RF) inductors.",

keywords = "Machine Learning, Deep Learning, Inductor, Coilcraft, Neural Network, LSTM, s-parameters, parameter prediction",

author = "Uttung Surange and Sourabh Khandelwal",

year = "2021",

doi = "10.1109/APMC52720.2021.9661689",

language = "English",

isbn = "9781665437837",

pages = "82--84",

booktitle = "2021 IEEE Asia-Pacific Microwave Conference (APMC)",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

address = "United States",

note = "2021 IEEE Asia-Pacific Microwave Conference, APMC 2021 ; Conference date: 28-11-2021 Through 01-12-2021",

}

Surange, U & Khandelwal, S 2021, L-Bot: a physically motivated deep learning based inductor modeling tool. in 2021 IEEE Asia-Pacific Microwave Conference (APMC). Institute of Electrical and Electronics Engineers (IEEE), Piscataway, NJ, pp. 82-84, 2021 IEEE Asia-Pacific Microwave Conference, APMC 2021, Virtual, Australia, 28/11/21. https://doi.org/10.1109/APMC52720.2021.9661689

L-Bot: a physically motivated deep learning based inductor modeling tool. / Surange, Uttung; Khandelwal, Sourabh.
2021 IEEE Asia-Pacific Microwave Conference (APMC). Piscataway, NJ: Institute of Electrical and Electronics Engineers (IEEE), 2021. p. 82-84.

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

TY - GEN

T1 - L-Bot

T2 - 2021 IEEE Asia-Pacific Microwave Conference, APMC 2021

AU - Surange, Uttung

AU - Khandelwal, Sourabh

PY - 2021

Y1 - 2021

N2 - In this paper, we present L-BOT, a physically motivated automatic inductor modeling tool based on deep learning algorithm. This tool uses the physical topology of inductor models representing the winding resistance, core losses, and capacitive effects in the inductor. The parameters of physical topology are automatically extracted using L-BOT after training the deep learning based engine. We show excellent accuracy of the inductor models generated from L-BOT for six different commercial radio-frequency (RF) inductors.

AB - In this paper, we present L-BOT, a physically motivated automatic inductor modeling tool based on deep learning algorithm. This tool uses the physical topology of inductor models representing the winding resistance, core losses, and capacitive effects in the inductor. The parameters of physical topology are automatically extracted using L-BOT after training the deep learning based engine. We show excellent accuracy of the inductor models generated from L-BOT for six different commercial radio-frequency (RF) inductors.

KW - Machine Learning

KW - Deep Learning

KW - Inductor

KW - Coilcraft

KW - Neural Network

KW - LSTM

KW - s-parameters

KW - parameter prediction

UR - http://www.scopus.com/inward/record.url?scp=85124402100&partnerID=8YFLogxK

U2 - 10.1109/APMC52720.2021.9661689

DO - 10.1109/APMC52720.2021.9661689

M3 - Conference proceeding contribution

AN - SCOPUS:85124402100

SN - 9781665437837

SP - 82

EP - 84

BT - 2021 IEEE Asia-Pacific Microwave Conference (APMC)

PB - Institute of Electrical and Electronics Engineers (IEEE)

CY - Piscataway, NJ

Y2 - 28 November 2021 through 1 December 2021

ER -

L-Bot: a physically motivated deep learning based inductor modeling tool

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this