Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and Application
A data-driven optimal control method for an air supply system in proton exchange membrane fuel cells (PEMFCs) is proposed with the aim of improving the PEMFC net output power and operational efficiency. Moreover, a marginal utility-based double-delay deep deterministic policy gradient (MU-4DPG) algo...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Frontiers Media S.A.
2021-09-01
|
Series: | Frontiers in Energy Research |
Subjects: | |
Online Access: | https://www.frontiersin.org/articles/10.3389/fenrg.2021.748782/full |
id |
doaj-12b048c0fd964f968c1b28ee30dfed4e |
---|---|
record_format |
Article |
spelling |
doaj-12b048c0fd964f968c1b28ee30dfed4e2021-09-30T05:07:27ZengFrontiers Media S.A.Frontiers in Energy Research2296-598X2021-09-01910.3389/fenrg.2021.748782748782Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and ApplicationJiawen Li0Kedong Zhu1Tao Yu2College of Electric Power, South China University of Technology, Guangzhou, ChinaChina Electric Power Research Institute (Nanjing), Nanjing, ChinaCollege of Electric Power, South China University of Technology, Guangzhou, ChinaA data-driven optimal control method for an air supply system in proton exchange membrane fuel cells (PEMFCs) is proposed with the aim of improving the PEMFC net output power and operational efficiency. Moreover, a marginal utility-based double-delay deep deterministic policy gradient (MU-4DPG) algorithm is proposed as a an offline tuner for the PID controller. The coefficients of the PID controller are rectified and optimized during training in order to enhance the controller’s performance. The design of the algorithm draws on the concept of marginal effects in Economics, in that the algorithm continuously switches between different forms of exploration noise during training so as to increase the diversity of samples, improve exploration efficiency and avoid Q-value overfitting, and ultimately improve the robustness of the algorithm. As detailed below, the effectiveness of the control method has been experimentally demonstrated.https://www.frontiersin.org/articles/10.3389/fenrg.2021.748782/fullproton exchange membrane fuel cellsPID controllerdata-driven optimal controldouble-delay deep deterministic policy gradientcontrol system |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Jiawen Li Kedong Zhu Tao Yu |
spellingShingle |
Jiawen Li Kedong Zhu Tao Yu Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and Application Frontiers in Energy Research proton exchange membrane fuel cells PID controller data-driven optimal control double-delay deep deterministic policy gradient control system |
author_facet |
Jiawen Li Kedong Zhu Tao Yu |
author_sort |
Jiawen Li |
title |
Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and Application |
title_short |
Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and Application |
title_full |
Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and Application |
title_fullStr |
Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and Application |
title_full_unstemmed |
Data-Driven Control for Proton Exchange Membrane Fuel Cells: Method and Application |
title_sort |
data-driven control for proton exchange membrane fuel cells: method and application |
publisher |
Frontiers Media S.A. |
series |
Frontiers in Energy Research |
issn |
2296-598X |
publishDate |
2021-09-01 |
description |
A data-driven optimal control method for an air supply system in proton exchange membrane fuel cells (PEMFCs) is proposed with the aim of improving the PEMFC net output power and operational efficiency. Moreover, a marginal utility-based double-delay deep deterministic policy gradient (MU-4DPG) algorithm is proposed as a an offline tuner for the PID controller. The coefficients of the PID controller are rectified and optimized during training in order to enhance the controller’s performance. The design of the algorithm draws on the concept of marginal effects in Economics, in that the algorithm continuously switches between different forms of exploration noise during training so as to increase the diversity of samples, improve exploration efficiency and avoid Q-value overfitting, and ultimately improve the robustness of the algorithm. As detailed below, the effectiveness of the control method has been experimentally demonstrated. |
topic |
proton exchange membrane fuel cells PID controller data-driven optimal control double-delay deep deterministic policy gradient control system |
url |
https://www.frontiersin.org/articles/10.3389/fenrg.2021.748782/full |
work_keys_str_mv |
AT jiawenli datadrivencontrolforprotonexchangemembranefuelcellsmethodandapplication AT kedongzhu datadrivencontrolforprotonexchangemembranefuelcellsmethodandapplication AT taoyu datadrivencontrolforprotonexchangemembranefuelcellsmethodandapplication |
_version_ |
1716863914643292160 |