Predictive Maintenance of Photovoltaic Panels via Deep Learning

Timo Huuhtanen; Alexander Jung

doi:10.1109/DSW.2018.8439898

Predictive Maintenance of Photovoltaic Panels via Deep Learning

Timo Huuhtanen, Alexander Jung

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

20 Citations (Scopus)

Abstract

We apply convolutional neural networks (CNN) for monitoring the operation of photovoltaic panels. In particular, we predict the daily electrical power curve of a photovoltaic panel based on the power curves of neighboring panels. An exceptionally large deviation between predicted and actual (observed) power curve can be used to indicate a malfunctioning panel. The problem is quite challenging because the power curve depends on many factors such as weather conditions and the surrounding objects (causing shadows with a regular time pattern). We demonstrate, by means of numerical experiments, that the proposed method is able to predict accurately the power curve of a functioning panel. Moreover, the proposed approach outperforms the existing approaches that are based on simple interpolation filters.

Original language	English
Title of host publication	2018 IEEE Data Science Workshop, DSW 2018 - Proceedings
Publisher	IEEE
Pages	66-70
Number of pages	5
ISBN (Print)	9781538644102
DOIs	https://doi.org/10.1109/DSW.2018.8439898
Publication status	Published - 17 Aug 2018
MoE publication type	A4 Article in a conference publication
Event	IEEE Data Science Workshop - Lausanne, Switzerland Duration: 4 Jun 2018 → 6 Jun 2018

Workshop

Workshop	IEEE Data Science Workshop
Abbreviated title	DSW
Country/Territory	Switzerland
City	Lausanne
Period	04/06/2018 → 06/06/2018

Keywords

Convolutional neural networks
machine learning
photovoltaic panels
predictive maintenance
wireless sensor networks

UN SDGs

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

Access to Document

10.1109/DSW.2018.8439898

SFX availability

Full text

Cite this

@inproceedings{7b94e8b3df584bfaba361a519f5c349c,

title = "Predictive Maintenance of Photovoltaic Panels via Deep Learning",

abstract = "We apply convolutional neural networks (CNN) for monitoring the operation of photovoltaic panels. In particular, we predict the daily electrical power curve of a photovoltaic panel based on the power curves of neighboring panels. An exceptionally large deviation between predicted and actual (observed) power curve can be used to indicate a malfunctioning panel. The problem is quite challenging because the power curve depends on many factors such as weather conditions and the surrounding objects (causing shadows with a regular time pattern). We demonstrate, by means of numerical experiments, that the proposed method is able to predict accurately the power curve of a functioning panel. Moreover, the proposed approach outperforms the existing approaches that are based on simple interpolation filters.",

keywords = "Convolutional neural networks, machine learning, photovoltaic panels, predictive maintenance, wireless sensor networks",

author = "Timo Huuhtanen and Alexander Jung",

year = "2018",

month = aug,

day = "17",

doi = "10.1109/DSW.2018.8439898",

language = "English",

isbn = "9781538644102",

pages = "66--70",

booktitle = "2018 IEEE Data Science Workshop, DSW 2018 - Proceedings",

publisher = "IEEE",

address = "United States",

note = "IEEE Data Science Workshop, DSW ; Conference date: 04-06-2018 Through 06-06-2018",

}

TY - GEN

T1 - Predictive Maintenance of Photovoltaic Panels via Deep Learning

AU - Huuhtanen, Timo

AU - Jung, Alexander

PY - 2018/8/17

Y1 - 2018/8/17

N2 - We apply convolutional neural networks (CNN) for monitoring the operation of photovoltaic panels. In particular, we predict the daily electrical power curve of a photovoltaic panel based on the power curves of neighboring panels. An exceptionally large deviation between predicted and actual (observed) power curve can be used to indicate a malfunctioning panel. The problem is quite challenging because the power curve depends on many factors such as weather conditions and the surrounding objects (causing shadows with a regular time pattern). We demonstrate, by means of numerical experiments, that the proposed method is able to predict accurately the power curve of a functioning panel. Moreover, the proposed approach outperforms the existing approaches that are based on simple interpolation filters.

AB - We apply convolutional neural networks (CNN) for monitoring the operation of photovoltaic panels. In particular, we predict the daily electrical power curve of a photovoltaic panel based on the power curves of neighboring panels. An exceptionally large deviation between predicted and actual (observed) power curve can be used to indicate a malfunctioning panel. The problem is quite challenging because the power curve depends on many factors such as weather conditions and the surrounding objects (causing shadows with a regular time pattern). We demonstrate, by means of numerical experiments, that the proposed method is able to predict accurately the power curve of a functioning panel. Moreover, the proposed approach outperforms the existing approaches that are based on simple interpolation filters.

KW - Convolutional neural networks

KW - machine learning

KW - photovoltaic panels

KW - predictive maintenance

KW - wireless sensor networks

UR - http://www.scopus.com/inward/record.url?scp=85053134200&partnerID=8YFLogxK

U2 - 10.1109/DSW.2018.8439898

DO - 10.1109/DSW.2018.8439898

M3 - Conference contribution

AN - SCOPUS:85053134200

SN - 9781538644102

SP - 66

EP - 70

BT - 2018 IEEE Data Science Workshop, DSW 2018 - Proceedings

PB - IEEE

T2 - IEEE Data Science Workshop

Y2 - 4 June 2018 through 6 June 2018

ER -

Predictive Maintenance of Photovoltaic Panels via Deep Learning

Abstract

Workshop

Keywords

UN SDGs

Access to Document

SFX availability

Other files and links

Fingerprint

Cite this