Modified inferential POD/ML for data-driven inverse procedure of steam reformer for 5-kW HT-PEMFC

被引：8

作者：

Koo, Bonchan ^{[1
]}

Jo, Taehyun ^{[2
]}

Lee, Dohyung ^{[3
]}

机构：

[1] Hanyang Univ, Mech Engn, Seoul, South Korea

[2] Hanyang Univ, Mech Design Engn, Seoul, South Korea

[3] Hanyang Univ, Mech Engn, Ansan Campus, Gyeonggi Do, South Korea

来源：

COMPUTERS & CHEMICAL ENGINEERING | 2019年 / 121卷

关键词：

Inverse problem; Machine learning; Proper orthogonal decomposition; Proton-exchange membrane fuel cells; Radial basis function network; Steam reformer; HEAT-CONDUCTION; METHANE;

D O I：

10.1016/j.compchemeng.2018.11.012

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

In this work, we applied and evaluated modified inferential proper orthogonal decomposition (POD)/machine learning (ML) to a steam reformer for 5-kW high-temperature proton-exchange membrane fuel cells (HT-PEMFC) involving heterogeneous chemical reactions, combustion, and fluid flow. The number of snapshots is limited by the inverse problem of a steam reformer yielding an intractable computational burden, and a limited number of snapshots and modes can yield unfavorable POD subspace projection results. In order to solve this problem, characteristic vectors are derived from the residual after POD projection and employed to the feature. We analyzed the details and distribution of the characteristic vector and investigated the extent of its influence on the inferential POD. Consequently, inferential POD/ML is improved by adding the characteristic vector of observation to the feature for ML. (C) 2018 Elsevier Ltd. All rights reserved.

引用

页码：375 / 387

页数：13