Iterative multi-task learning for time-series modeling of solar panel PV outputs

Tahasin Shireen, Chenhui Shao, Hui Wang, Jingjing Li, Xi Zhang, Mingyang Li

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Time-series modeling of PV output for solar panels can help solar panel owners understand the power systems’ time-varying behavior and be prepared for the load demand. The time-series forecast/prediction can become challenging due to many missing observations or a lack of historical records that are not sufficient to establish statistical models. Increasing PV measurement frequency over a longer period increases the cost in the detection of the PV fluctuation. This paper proposes an efficient approach to iterative multi-task learning for time series (MTL-GP-TS) that improves prediction of the PV output without increasing measurement efforts by sharing the information among PV data from multiple similar solar panels. The proposed iterative MTL-GP-TS model learns/imputes unobserved or missing values in a dataset of time series associated with the solar panel of interest to predict the PV trend. Additionally, the method improves and generalizes the traditional multi-task learning for Gaussian Process to the learning of both global trend and local irregular components in time series. A real-world case study demonstrated that the proposed method could result in substantial improvement of predictions over conventional approaches. The paper also discusses the selection of parameters and data sources when implementing the proposed algorithm.

Original languageEnglish (US)
Pages (from-to)654-662
Number of pages9
JournalApplied Energy
Volume212
DOIs
StatePublished - Feb 15 2018

Fingerprint

Time series
learning
time series
modeling
prediction
Time varying systems
historical record
cost
Costs
trend
method

All Science Journal Classification (ASJC) codes

  • Building and Construction
  • Energy(all)
  • Mechanical Engineering
  • Management, Monitoring, Policy and Law

Cite this

Shireen, Tahasin ; Shao, Chenhui ; Wang, Hui ; Li, Jingjing ; Zhang, Xi ; Li, Mingyang. / Iterative multi-task learning for time-series modeling of solar panel PV outputs. In: Applied Energy. 2018 ; Vol. 212. pp. 654-662.
@article{bdb5e6d3519948be9804025171e77306,
title = "Iterative multi-task learning for time-series modeling of solar panel PV outputs",
abstract = "Time-series modeling of PV output for solar panels can help solar panel owners understand the power systems’ time-varying behavior and be prepared for the load demand. The time-series forecast/prediction can become challenging due to many missing observations or a lack of historical records that are not sufficient to establish statistical models. Increasing PV measurement frequency over a longer period increases the cost in the detection of the PV fluctuation. This paper proposes an efficient approach to iterative multi-task learning for time series (MTL-GP-TS) that improves prediction of the PV output without increasing measurement efforts by sharing the information among PV data from multiple similar solar panels. The proposed iterative MTL-GP-TS model learns/imputes unobserved or missing values in a dataset of time series associated with the solar panel of interest to predict the PV trend. Additionally, the method improves and generalizes the traditional multi-task learning for Gaussian Process to the learning of both global trend and local irregular components in time series. A real-world case study demonstrated that the proposed method could result in substantial improvement of predictions over conventional approaches. The paper also discusses the selection of parameters and data sources when implementing the proposed algorithm.",
author = "Tahasin Shireen and Chenhui Shao and Hui Wang and Jingjing Li and Xi Zhang and Mingyang Li",
year = "2018",
month = "2",
day = "15",
doi = "10.1016/j.apenergy.2017.12.058",
language = "English (US)",
volume = "212",
pages = "654--662",
journal = "Applied Energy",
issn = "0306-2619",
publisher = "Elsevier BV",

}

Iterative multi-task learning for time-series modeling of solar panel PV outputs. / Shireen, Tahasin; Shao, Chenhui; Wang, Hui; Li, Jingjing; Zhang, Xi; Li, Mingyang.

In: Applied Energy, Vol. 212, 15.02.2018, p. 654-662.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Iterative multi-task learning for time-series modeling of solar panel PV outputs

AU - Shireen, Tahasin

AU - Shao, Chenhui

AU - Wang, Hui

AU - Li, Jingjing

AU - Zhang, Xi

AU - Li, Mingyang

PY - 2018/2/15

Y1 - 2018/2/15

N2 - Time-series modeling of PV output for solar panels can help solar panel owners understand the power systems’ time-varying behavior and be prepared for the load demand. The time-series forecast/prediction can become challenging due to many missing observations or a lack of historical records that are not sufficient to establish statistical models. Increasing PV measurement frequency over a longer period increases the cost in the detection of the PV fluctuation. This paper proposes an efficient approach to iterative multi-task learning for time series (MTL-GP-TS) that improves prediction of the PV output without increasing measurement efforts by sharing the information among PV data from multiple similar solar panels. The proposed iterative MTL-GP-TS model learns/imputes unobserved or missing values in a dataset of time series associated with the solar panel of interest to predict the PV trend. Additionally, the method improves and generalizes the traditional multi-task learning for Gaussian Process to the learning of both global trend and local irregular components in time series. A real-world case study demonstrated that the proposed method could result in substantial improvement of predictions over conventional approaches. The paper also discusses the selection of parameters and data sources when implementing the proposed algorithm.

AB - Time-series modeling of PV output for solar panels can help solar panel owners understand the power systems’ time-varying behavior and be prepared for the load demand. The time-series forecast/prediction can become challenging due to many missing observations or a lack of historical records that are not sufficient to establish statistical models. Increasing PV measurement frequency over a longer period increases the cost in the detection of the PV fluctuation. This paper proposes an efficient approach to iterative multi-task learning for time series (MTL-GP-TS) that improves prediction of the PV output without increasing measurement efforts by sharing the information among PV data from multiple similar solar panels. The proposed iterative MTL-GP-TS model learns/imputes unobserved or missing values in a dataset of time series associated with the solar panel of interest to predict the PV trend. Additionally, the method improves and generalizes the traditional multi-task learning for Gaussian Process to the learning of both global trend and local irregular components in time series. A real-world case study demonstrated that the proposed method could result in substantial improvement of predictions over conventional approaches. The paper also discusses the selection of parameters and data sources when implementing the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=85038106127&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038106127&partnerID=8YFLogxK

U2 - 10.1016/j.apenergy.2017.12.058

DO - 10.1016/j.apenergy.2017.12.058

M3 - Article

AN - SCOPUS:85038106127

VL - 212

SP - 654

EP - 662

JO - Applied Energy

JF - Applied Energy

SN - 0306-2619

ER -