Comparison and sensitivity analysis of methods for solar PV power prediction

Mashud Rana; Ashfaqur Rahman; Liwan Liyanage; Mohammed Nazim Uddin

doi:10.1007/978-3-030-04503-6_32

Comparison and sensitivity analysis of methods for solar PV power prediction

Mashud Rana
, Ashfaqur Rahman
, Liwan Liyanage
, Mohammed Nazim Uddin

Research output: Chapter in Book / Conference Paper › Conference Paper › peer-review

5 Citations (Scopus)

Abstract

The variable nature of solar power output from PhotoVoltaic (PV) systems is the main obstacle for penetration of such power into the electricity grid. Thus, numerous methods have been proposed in the literature to construct forecasting models. In this paper, we present a comprehensive comparison of a set of prominent methods that utilize weather prediction for future. Firstly, we evaluate the prediction accuracy of widely used Neural Network (NN), Support Vector Regression (SVR), k-Nearest Neighbour (kNN), Multiple Linear Regression (MLR), and two persistent methods using four data sets for 2 years. We then analyse the sensitivities of their prediction accuracy to 1"”25% possible error in the future weather prediction obtained from the Bureau of Meteorology (BoM). Results demonstrate that ensemble of NNs is the most promising method and achieves substantial improvement in accuracy over other prediction methods.

Original language	English
Title of host publication	Trends and Applications in Knowledge Discovery and Data Mining: PAKDD 2018 Workshops, Melbourne, Vic, Australia, June 3 2018, Revised Selected Papers
Publisher	Springer
Pages	333-344
Number of pages	12
ISBN (Print)	9783030045029
DOIs	https://doi.org/10.1007/978-3-030-04503-6_32
Publication status	Published - 2018
Event	Pacific-Asia Conference on Knowledge Discovery and Data Mining - Duration: 3 Jun 2018 → …

Publication series

Name
ISSN (Print)	0302-9743

Conference

Conference	Pacific-Asia Conference on Knowledge Discovery and Data Mining
Period	3/06/18 → …

UN SDGs

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

SDG 7 Affordable and Clean Energy

Keywords

nearest neighbor analysis (statistics)
neural networks (computer science)
regression analysis
solar energy
vector analysis
weather forecasting

Access to Document

10.1007/978-3-030-04503-6_32

Cite this

@inproceedings{a66e3b1084ea437293ad2a498cdbd8de,

title = "Comparison and sensitivity analysis of methods for solar PV power prediction",

abstract = "The variable nature of solar power output from PhotoVoltaic (PV) systems is the main obstacle for penetration of such power into the electricity grid. Thus, numerous methods have been proposed in the literature to construct forecasting models. In this paper, we present a comprehensive comparison of a set of prominent methods that utilize weather prediction for future. Firstly, we evaluate the prediction accuracy of widely used Neural Network (NN), Support Vector Regression (SVR), k-Nearest Neighbour (kNN), Multiple Linear Regression (MLR), and two persistent methods using four data sets for 2 years. We then analyse the sensitivities of their prediction accuracy to 1{"}”25\% possible error in the future weather prediction obtained from the Bureau of Meteorology (BoM). Results demonstrate that ensemble of NNs is the most promising method and achieves substantial improvement in accuracy over other prediction methods.",

keywords = "nearest neighbor analysis (statistics), neural networks (computer science), regression analysis, solar energy, vector analysis, weather forecasting",

author = "Mashud Rana and Ashfaqur Rahman and Liwan Liyanage and Uddin, \{Mohammed Nazim\}",

year = "2018",

doi = "10.1007/978-3-030-04503-6\_32",

language = "English",

isbn = "9783030045029",

publisher = "Springer",

pages = "333--344",

booktitle = "Trends and Applications in Knowledge Discovery and Data Mining: PAKDD 2018 Workshops, Melbourne, Vic, Australia, June 3 2018, Revised Selected Papers",

note = "Pacific-Asia Conference on Knowledge Discovery and Data Mining ; Conference date: 03-06-2018",

}

Rana, M, Rahman, A, Liyanage, L & Uddin, MN 2018, Comparison and sensitivity analysis of methods for solar PV power prediction. in Trends and Applications in Knowledge Discovery and Data Mining: PAKDD 2018 Workshops, Melbourne, Vic, Australia, June 3 2018, Revised Selected Papers. Springer, pp. 333-344, Pacific-Asia Conference on Knowledge Discovery and Data Mining, 3/06/18. https://doi.org/10.1007/978-3-030-04503-6_32

Comparison and sensitivity analysis of methods for solar PV power prediction. / Rana, Mashud; Rahman, Ashfaqur; Liyanage, Liwan et al.
Trends and Applications in Knowledge Discovery and Data Mining: PAKDD 2018 Workshops, Melbourne, Vic, Australia, June 3 2018, Revised Selected Papers. Springer, 2018. p. 333-344.

Research output: Chapter in Book / Conference Paper › Conference Paper › peer-review

TY - GEN

T1 - Comparison and sensitivity analysis of methods for solar PV power prediction

AU - Rana, Mashud

AU - Rahman, Ashfaqur

AU - Liyanage, Liwan

AU - Uddin, Mohammed Nazim

PY - 2018

Y1 - 2018

N2 - The variable nature of solar power output from PhotoVoltaic (PV) systems is the main obstacle for penetration of such power into the electricity grid. Thus, numerous methods have been proposed in the literature to construct forecasting models. In this paper, we present a comprehensive comparison of a set of prominent methods that utilize weather prediction for future. Firstly, we evaluate the prediction accuracy of widely used Neural Network (NN), Support Vector Regression (SVR), k-Nearest Neighbour (kNN), Multiple Linear Regression (MLR), and two persistent methods using four data sets for 2 years. We then analyse the sensitivities of their prediction accuracy to 1"”25% possible error in the future weather prediction obtained from the Bureau of Meteorology (BoM). Results demonstrate that ensemble of NNs is the most promising method and achieves substantial improvement in accuracy over other prediction methods.

AB - The variable nature of solar power output from PhotoVoltaic (PV) systems is the main obstacle for penetration of such power into the electricity grid. Thus, numerous methods have been proposed in the literature to construct forecasting models. In this paper, we present a comprehensive comparison of a set of prominent methods that utilize weather prediction for future. Firstly, we evaluate the prediction accuracy of widely used Neural Network (NN), Support Vector Regression (SVR), k-Nearest Neighbour (kNN), Multiple Linear Regression (MLR), and two persistent methods using four data sets for 2 years. We then analyse the sensitivities of their prediction accuracy to 1"”25% possible error in the future weather prediction obtained from the Bureau of Meteorology (BoM). Results demonstrate that ensemble of NNs is the most promising method and achieves substantial improvement in accuracy over other prediction methods.

KW - nearest neighbor analysis (statistics)

KW - neural networks (computer science)

KW - regression analysis

KW - solar energy

KW - vector analysis

KW - weather forecasting

UR - http://hdl.handle.net/1959.7/uws:52486

U2 - 10.1007/978-3-030-04503-6_32

DO - 10.1007/978-3-030-04503-6_32

M3 - Conference Paper

SN - 9783030045029

SP - 333

EP - 344

BT - Trends and Applications in Knowledge Discovery and Data Mining: PAKDD 2018 Workshops, Melbourne, Vic, Australia, June 3 2018, Revised Selected Papers

PB - Springer

T2 - Pacific-Asia Conference on Knowledge Discovery and Data Mining

Y2 - 3 June 2018

ER -

Comparison and sensitivity analysis of methods for solar PV power prediction

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this