Fault diagnosis for the planetary gearbox based on an improved LightGBM

Siyuan Zhang; Yang Liu

doi:10.1109/SAFEPROCESS52771.2021.9693594

Fault diagnosis for the planetary gearbox based on an improved LightGBM

Siyuan Zhang, Yang Liu

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

1 Citation (Scopus)

Abstract

Planetary gearboxes are usually regarded as a key part of rotating machinery, so fault diagnosis of planetary gearboxes is significant. In this paper, an improved Light Gradient Boosting Machine (LightGBM) is proposed for fault diagnosis of planetary gearboxes. The vibration signal data of various fault modes are obtained with the experimental platform, and the simple feature extraction and normalization are carried out. LightGBM is employed to classify the processed data and identify the faulty data. Through 5-fold cross-validation, the Bayesian Optimization Algorithm (BOA) is adopted to determine the optimal combination of the hyper-parameters which are decisive in the final classification performance of the LightGBM. The comparative experiments with other methods have shown the superiority of the proposed method in fault diagnosis of the planetary gearbox.

Original language	English
Title of host publication	2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes
Subtitle of host publication	SAFEPROCESS 2021
Publisher	Institute of Electrical and Electronics Engineers Inc.
Number of pages	6
ISBN (Electronic)	9781665401159
ISBN (Print)	9781665401166
DOIs	https://doi.org/10.1109/SAFEPROCESS52771.2021.9693594
Publication status	Published - 1 Feb 2022
Externally published	Yes
Event	12th CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes - Chengdu, China Duration: 17 Dec 2021 → 18 Dec 2021 Conference number: 12 https://fdd2021.aconf.org/

Conference

Conference	12th CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes
Abbreviated title	CAA SAFEPROCESS 2021
Country/Territory	China
City	Chengdu
Period	17/12/21 → 18/12/21
Internet address	https://fdd2021.aconf.org/

Access to Document

10.1109/SAFEPROCESS52771.2021.9693594Licence: Unspecified

Cite this

@inproceedings{5df955ef714741d7bb7c5bef3d8a9f2c,

title = "Fault diagnosis for the planetary gearbox based on an improved LightGBM",

abstract = "Planetary gearboxes are usually regarded as a key part of rotating machinery, so fault diagnosis of planetary gearboxes is significant. In this paper, an improved Light Gradient Boosting Machine (LightGBM) is proposed for fault diagnosis of planetary gearboxes. The vibration signal data of various fault modes are obtained with the experimental platform, and the simple feature extraction and normalization are carried out. LightGBM is employed to classify the processed data and identify the faulty data. Through 5-fold cross-validation, the Bayesian Optimization Algorithm (BOA) is adopted to determine the optimal combination of the hyper-parameters which are decisive in the final classification performance of the LightGBM. The comparative experiments with other methods have shown the superiority of the proposed method in fault diagnosis of the planetary gearbox.",

keywords = "Bayesian Optimization Algorithm, feature extraction, LightGBM, normalization, Planetary gearboxes",

author = "Siyuan Zhang and Yang Liu",

note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 12th CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes, CAA SAFEPROCESS 2021 ; Conference date: 17-12-2021 Through 18-12-2021",

year = "2022",

month = feb,

day = "1",

doi = "10.1109/SAFEPROCESS52771.2021.9693594",

language = "English",

isbn = "9781665401166",

booktitle = "2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

url = "https://fdd2021.aconf.org/",

}

Zhang, S & Liu, Y 2022, Fault diagnosis for the planetary gearbox based on an improved LightGBM. in 2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes: SAFEPROCESS 2021., 9693594, Institute of Electrical and Electronics Engineers Inc., 12th CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes, Chengdu, China, 17/12/21. https://doi.org/10.1109/SAFEPROCESS52771.2021.9693594

Fault diagnosis for the planetary gearbox based on an improved LightGBM. / Zhang, Siyuan; Liu, Yang.

2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes: SAFEPROCESS 2021. Institute of Electrical and Electronics Engineers Inc., 2022. 9693594.

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

TY - GEN

T1 - Fault diagnosis for the planetary gearbox based on an improved LightGBM

AU - Zhang, Siyuan

AU - Liu, Yang

N1 - Conference code: 12

PY - 2022/2/1

Y1 - 2022/2/1

N2 - Planetary gearboxes are usually regarded as a key part of rotating machinery, so fault diagnosis of planetary gearboxes is significant. In this paper, an improved Light Gradient Boosting Machine (LightGBM) is proposed for fault diagnosis of planetary gearboxes. The vibration signal data of various fault modes are obtained with the experimental platform, and the simple feature extraction and normalization are carried out. LightGBM is employed to classify the processed data and identify the faulty data. Through 5-fold cross-validation, the Bayesian Optimization Algorithm (BOA) is adopted to determine the optimal combination of the hyper-parameters which are decisive in the final classification performance of the LightGBM. The comparative experiments with other methods have shown the superiority of the proposed method in fault diagnosis of the planetary gearbox.

AB - Planetary gearboxes are usually regarded as a key part of rotating machinery, so fault diagnosis of planetary gearboxes is significant. In this paper, an improved Light Gradient Boosting Machine (LightGBM) is proposed for fault diagnosis of planetary gearboxes. The vibration signal data of various fault modes are obtained with the experimental platform, and the simple feature extraction and normalization are carried out. LightGBM is employed to classify the processed data and identify the faulty data. Through 5-fold cross-validation, the Bayesian Optimization Algorithm (BOA) is adopted to determine the optimal combination of the hyper-parameters which are decisive in the final classification performance of the LightGBM. The comparative experiments with other methods have shown the superiority of the proposed method in fault diagnosis of the planetary gearbox.

KW - Bayesian Optimization Algorithm

KW - feature extraction

KW - LightGBM

KW - normalization

KW - Planetary gearboxes

UR - http://www.scopus.com/inward/record.url?scp=85126901438&partnerID=8YFLogxK

U2 - 10.1109/SAFEPROCESS52771.2021.9693594

DO - 10.1109/SAFEPROCESS52771.2021.9693594

M3 - Conference contribution

AN - SCOPUS:85126901438

SN - 9781665401166

BT - 2021 CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 12th CAA Symposium on Fault Detection, Supervision, and Safety for Technical Processes

Y2 - 17 December 2021 through 18 December 2021

ER -

Fault diagnosis for the planetary gearbox based on an improved LightGBM

Abstract

Conference

Access to Document

Other files and links

Cite this