Validation of neural network-based fault diagnosis for multi-stack fuel cell systems:  Stack voltage deviation detection

Antti Pohjoranta; Marco Sorrentino; Cesare Pianese; F. Amatruda; Tero Hottinen

doi:10.1016/j.egypro.2015.12.072

Validation of neural network-based fault diagnosis for multi-stack fuel cell systems: Stack voltage deviation detection

Antti Pohjoranta, Marco Sorrentino (Corresponding Author), Cesare Pianese, F. Amatruda, Tero Hottinen

Research output: Contribution to journal › Article › Scientific › peer-review

18 Citations (Scopus)

Abstract

This paper presents (i) an algorithm for the detection of unexpected stack voltage deviations in an Solid Oxide Fuel Cells (SOFC)-based power system with multiple stacks and (ii) its validation in a simulated online environment. The algorithm is based on recurrent neural networks (RNNs) and is validated by using operating data from the Wärtsilä WFC20 multi-stack SOFC system. The voltage deviation detection is based on statistical testing. Instead of a hardware implementation in the actual power plant, the algorithm is validated in a simulated online environment that provides data I/O communication based on the OPC (i.e. Object Linking and Embedding (OLE) for Process Control) protocol, which is also the technology utilized in the real hardware environment. The validation tests show that the RNN-based algorithm effectively detects unwanted stack voltage deviations and also that it is online-viable.

Original language	English
Pages (from-to)	173-181
Journal	Energy Procedia
Volume	81
DOIs	https://doi.org/10.1016/j.egypro.2015.12.072
Publication status	Published - 2015
MoE publication type	A1 Journal article-refereed
Event	69th Conference of the Italian Thermal Machines Engineering Association, ATI 2014 - Milan, Italy Duration: 10 Sept 2014 → 12 Sept 2014

Keywords

solid oxide fuel cell
multi-stack system
voltage fault diagnosis
online validation
neural networks

Access to Document

10.1016/j.egypro.2015.12.072

Cite this

@article{e448675dc9a84bfeb322571316256077,

title = "Validation of neural network-based fault diagnosis for multi-stack fuel cell systems: Stack voltage deviation detection",

abstract = "This paper presents (i) an algorithm for the detection of unexpected stack voltage deviations in an Solid Oxide Fuel Cells (SOFC)-based power system with multiple stacks and (ii) its validation in a simulated online environment. The algorithm is based on recurrent neural networks (RNNs) and is validated by using operating data from the W{\"a}rtsil{\"a} WFC20 multi-stack SOFC system. The voltage deviation detection is based on statistical testing. Instead of a hardware implementation in the actual power plant, the algorithm is validated in a simulated online environment that provides data I/O communication based on the OPC (i.e. Object Linking and Embedding (OLE) for Process Control) protocol, which is also the technology utilized in the real hardware environment. The validation tests show that the RNN-based algorithm effectively detects unwanted stack voltage deviations and also that it is online-viable.",

keywords = "solid oxide fuel cell, multi-stack system, voltage fault diagnosis, online validation, neural networks",

author = "Antti Pohjoranta and Marco Sorrentino and Cesare Pianese and F. Amatruda and Tero Hottinen",

year = "2015",

doi = "10.1016/j.egypro.2015.12.072",

language = "English",

volume = "81",

pages = "173--181",

journal = "Energy Procedia",

issn = "1876-6102",

publisher = "Elsevier",

note = "69th Conference of the Italian Thermal Machines Engineering Association, ATI 2014 ; Conference date: 10-09-2014 Through 12-09-2014",

}

TY - JOUR

T1 - Validation of neural network-based fault diagnosis for multi-stack fuel cell systems

T2 - 69th Conference of the Italian Thermal Machines Engineering Association, ATI 2014

AU - Pohjoranta, Antti

AU - Sorrentino, Marco

AU - Pianese, Cesare

AU - Amatruda, F.

AU - Hottinen, Tero

PY - 2015

Y1 - 2015

N2 - This paper presents (i) an algorithm for the detection of unexpected stack voltage deviations in an Solid Oxide Fuel Cells (SOFC)-based power system with multiple stacks and (ii) its validation in a simulated online environment. The algorithm is based on recurrent neural networks (RNNs) and is validated by using operating data from the Wärtsilä WFC20 multi-stack SOFC system. The voltage deviation detection is based on statistical testing. Instead of a hardware implementation in the actual power plant, the algorithm is validated in a simulated online environment that provides data I/O communication based on the OPC (i.e. Object Linking and Embedding (OLE) for Process Control) protocol, which is also the technology utilized in the real hardware environment. The validation tests show that the RNN-based algorithm effectively detects unwanted stack voltage deviations and also that it is online-viable.

AB - This paper presents (i) an algorithm for the detection of unexpected stack voltage deviations in an Solid Oxide Fuel Cells (SOFC)-based power system with multiple stacks and (ii) its validation in a simulated online environment. The algorithm is based on recurrent neural networks (RNNs) and is validated by using operating data from the Wärtsilä WFC20 multi-stack SOFC system. The voltage deviation detection is based on statistical testing. Instead of a hardware implementation in the actual power plant, the algorithm is validated in a simulated online environment that provides data I/O communication based on the OPC (i.e. Object Linking and Embedding (OLE) for Process Control) protocol, which is also the technology utilized in the real hardware environment. The validation tests show that the RNN-based algorithm effectively detects unwanted stack voltage deviations and also that it is online-viable.

KW - solid oxide fuel cell

KW - multi-stack system

KW - voltage fault diagnosis

KW - online validation

KW - neural networks

UR - https://www.sciencedirect.com/journal/energy-procedia/vol/81/suppl/C

U2 - 10.1016/j.egypro.2015.12.072

DO - 10.1016/j.egypro.2015.12.072

M3 - Article

SN - 1876-6102

VL - 81

SP - 173

EP - 181

JO - Energy Procedia

JF - Energy Procedia

Y2 - 10 September 2014 through 12 September 2014

ER -

Validation of neural network-based fault diagnosis for multi-stack fuel cell systems: Stack voltage deviation detection

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this