A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network

Yongtang Ye; Wenbo Zhang; Kunsong Lin; Ping Xu; Yunxia Chen

doi:10.1109/PHM-Shanghai49105.2020.9280989

A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network

Yongtang Ye
, Wenbo Zhang
, Kunsong Lin
, Ping Xu
, Yunxia Chen

School of Automation Science and Electrical Engineering

Beihang University

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

Abstract

Support Vector Regression (SVR) is extensively used in various engineering fields. Hyperparameters directly affect the prediction accuracy of SVR model. Traditional hyperparameters identification method simply relies on data of the current sample, which is prone to overfitting with a low prediction accuracy. Therefore, a new hyperparameters identification method is put forward in this paper based on Back Propagation (BP) neural network. Firstly, the optimum hyperparameters of SVR model for each sample in historical database are obtained by a new optimization model proposed. Secondly, the mapping relationship between input data and optimum hyperparameters is built on the basis of BP neural network using multiple historical samples. Furthermore, the optimum hyperparameters for a new sample can be directly acquired from the neural network established. Finally, a case study on state of health estimation for lithium-ion batteries applying SVR is conducted to verify the effectiveness of the presented method.

Original language	English
Title of host publication	2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020
Editors	Wei Guo, Steven Li, Qiang Miao
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728159454
DOIs	https://doi.org/10.1109/PHM-Shanghai49105.2020.9280989
State	Published - 16 Oct 2020
Event	2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020 - Shanghai, China Duration: 16 Oct 2020 → 18 Oct 2020

Publication series

Name	2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020

Conference

Conference	2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020
Country/Territory	China
City	Shanghai
Period	16/10/20 → 18/10/20

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

Back Propagation Neural Network
Hyperparameters Identification
State of Health Estimation
Support Vector Regression

Access to Document

10.1109/PHM-Shanghai49105.2020.9280989

Cite this

Ye, Y., Zhang, W., Lin, K., Xu, P., & Chen, Y. (2020). A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network. In W. Guo, S. Li, & Q. Miao (Eds.), 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020 Article 9280989 (2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PHM-Shanghai49105.2020.9280989

Ye, Yongtang ; Zhang, Wenbo ; Lin, Kunsong et al. / A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network. 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020. editor / Wei Guo ; Steven Li ; Qiang Miao. Institute of Electrical and Electronics Engineers Inc., 2020. (2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020).

@inproceedings{0f730e3e3d904413a06aa732e651f5ec,

title = "A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network",

abstract = "Support Vector Regression (SVR) is extensively used in various engineering fields. Hyperparameters directly affect the prediction accuracy of SVR model. Traditional hyperparameters identification method simply relies on data of the current sample, which is prone to overfitting with a low prediction accuracy. Therefore, a new hyperparameters identification method is put forward in this paper based on Back Propagation (BP) neural network. Firstly, the optimum hyperparameters of SVR model for each sample in historical database are obtained by a new optimization model proposed. Secondly, the mapping relationship between input data and optimum hyperparameters is built on the basis of BP neural network using multiple historical samples. Furthermore, the optimum hyperparameters for a new sample can be directly acquired from the neural network established. Finally, a case study on state of health estimation for lithium-ion batteries applying SVR is conducted to verify the effectiveness of the presented method.",

keywords = "Back Propagation Neural Network, Hyperparameters Identification, State of Health Estimation, Support Vector Regression",

author = "Yongtang Ye and Wenbo Zhang and Kunsong Lin and Ping Xu and Yunxia Chen",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020 ; Conference date: 16-10-2020 Through 18-10-2020",

year = "2020",

month = oct,

day = "16",

doi = "10.1109/PHM-Shanghai49105.2020.9280989",

language = "英语",

series = "2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020",

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

editor = "Wei Guo and Steven Li and Qiang Miao",

booktitle = "2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020",

address = "美国",

}

Ye, Y, Zhang, W, Lin, K, Xu, P & Chen, Y 2020, A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network. in W Guo, S Li & Q Miao (eds), 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020., 9280989, 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020, Institute of Electrical and Electronics Engineers Inc., 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020, Shanghai, China, 16/10/20. https://doi.org/10.1109/PHM-Shanghai49105.2020.9280989

A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network. / Ye, Yongtang; Zhang, Wenbo; Lin, Kunsong et al.
2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020. ed. / Wei Guo; Steven Li; Qiang Miao. Institute of Electrical and Electronics Engineers Inc., 2020. 9280989 (2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020).

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

TY - GEN

T1 - A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network

AU - Ye, Yongtang

AU - Zhang, Wenbo

AU - Lin, Kunsong

AU - Xu, Ping

AU - Chen, Yunxia

PY - 2020/10/16

Y1 - 2020/10/16

N2 - Support Vector Regression (SVR) is extensively used in various engineering fields. Hyperparameters directly affect the prediction accuracy of SVR model. Traditional hyperparameters identification method simply relies on data of the current sample, which is prone to overfitting with a low prediction accuracy. Therefore, a new hyperparameters identification method is put forward in this paper based on Back Propagation (BP) neural network. Firstly, the optimum hyperparameters of SVR model for each sample in historical database are obtained by a new optimization model proposed. Secondly, the mapping relationship between input data and optimum hyperparameters is built on the basis of BP neural network using multiple historical samples. Furthermore, the optimum hyperparameters for a new sample can be directly acquired from the neural network established. Finally, a case study on state of health estimation for lithium-ion batteries applying SVR is conducted to verify the effectiveness of the presented method.

AB - Support Vector Regression (SVR) is extensively used in various engineering fields. Hyperparameters directly affect the prediction accuracy of SVR model. Traditional hyperparameters identification method simply relies on data of the current sample, which is prone to overfitting with a low prediction accuracy. Therefore, a new hyperparameters identification method is put forward in this paper based on Back Propagation (BP) neural network. Firstly, the optimum hyperparameters of SVR model for each sample in historical database are obtained by a new optimization model proposed. Secondly, the mapping relationship between input data and optimum hyperparameters is built on the basis of BP neural network using multiple historical samples. Furthermore, the optimum hyperparameters for a new sample can be directly acquired from the neural network established. Finally, a case study on state of health estimation for lithium-ion batteries applying SVR is conducted to verify the effectiveness of the presented method.

KW - Back Propagation Neural Network

KW - Hyperparameters Identification

KW - State of Health Estimation

KW - Support Vector Regression

UR - https://www.scopus.com/pages/publications/85099708499

U2 - 10.1109/PHM-Shanghai49105.2020.9280989

DO - 10.1109/PHM-Shanghai49105.2020.9280989

M3 - 会议稿件

AN - SCOPUS:85099708499

T3 - 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020

BT - 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020

A2 - Guo, Wei

A2 - Li, Steven

A2 - Miao, Qiang

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020

Y2 - 16 October 2020 through 18 October 2020

ER -

Ye Y, Zhang W, Lin K, Xu P, Chen Y. A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network. In Guo W, Li S, Miao Q, editors, 2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020. Institute of Electrical and Electronics Engineers Inc. 2020. 9280989. (2020 Global Reliability and Prognostics and Health Management, PHM-Shanghai 2020). doi: 10.1109/PHM-Shanghai49105.2020.9280989

A New Hyperparameters Identification Method of Support Vector Regression Model Based on Back Propagation Neural Network

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this