Day-ahead electricity prices forecasting using artificial neural networks

Qi Tang; Danzhen Gu

Day-ahead electricity prices forecasting using artificial neural networks

Qi Tang, Danzhen Gu

The University of Manchester

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

Abstract

This paper presents specifically how artificial neural network (ANN) is applied to forecast day-ahead electricity price in the deregulated electricity market. Market clearing price (MCP) forecasting has been more and more significant in new restructured market because both generating companies and consumers rely on it to prepare their bidding strategies and expect to maximize respective benefits with low risks. But the prediction of MCP is complex because various uncertainties interact in an intricate way. Hence, ANN is proposed to solve high complicated nonlinear problems like MCP forecasting due to its powerful capability of learning mechanism if enough data for training. The paper focuses on influences brought by different input architectures with different training methods and proposes a basic optimal ANN architecture for day-ahead MCP forecasting. © 2009 IEEE.

Original language	English
Title of host publication	2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009\|Int. Conf. Artif. Intell. Comput. Intell., AICI
Pages	511-514
Number of pages	3
Volume	2
Publication status	Published - 2009
Event	2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009 - Shanghai Duration: 1 Jul 2009 → … http://dx.doi.org/10.1109/AICI.2009.184

Conference

Conference	2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009
City	Shanghai
Period	1/07/09 → …
Internet address	http://dx.doi.org/10.1109/AICI.2009.184

Keywords

Artificial neural network
Forecasting
Market clearing price

Access to Document

Cite this

@inproceedings{ccbc8ed34519416a8dd9163ebf456912,

title = "Day-ahead electricity prices forecasting using artificial neural networks",

abstract = "This paper presents specifically how artificial neural network (ANN) is applied to forecast day-ahead electricity price in the deregulated electricity market. Market clearing price (MCP) forecasting has been more and more significant in new restructured market because both generating companies and consumers rely on it to prepare their bidding strategies and expect to maximize respective benefits with low risks. But the prediction of MCP is complex because various uncertainties interact in an intricate way. Hence, ANN is proposed to solve high complicated nonlinear problems like MCP forecasting due to its powerful capability of learning mechanism if enough data for training. The paper focuses on influences brought by different input architectures with different training methods and proposes a basic optimal ANN architecture for day-ahead MCP forecasting. {\textcopyright} 2009 IEEE.",

keywords = "Artificial neural network, Forecasting, Market clearing price",

author = "Qi Tang and Danzhen Gu",

note = "Artificial neural network;Bidding strategy;Deregulated electricity market;Electricity prices;Electricity prices forecasting;Generating companies;Learning mechanism;Market Clearing Price;Market clearing prices;Nonlinear problems;Training methods;; 2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009 ; Conference date: 01-07-2009",

year = "2009",

language = "English",

isbn = "9780769538167",

volume = "2",

pages = "511--514",

booktitle = "2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009|Int. Conf. Artif. Intell. Comput. Intell., AICI",

url = "http://dx.doi.org/10.1109/AICI.2009.184",

}

Tang, Q & Gu, D 2009, Day-ahead electricity prices forecasting using artificial neural networks. in 2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009|Int. Conf. Artif. Intell. Comput. Intell., AICI. vol. 2, pp. 511-514, 2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009, Shanghai, 1/07/09. <http://dx.doi.org/10.1109/AICI.2009.184>

Day-ahead electricity prices forecasting using artificial neural networks. / Tang, Qi; Gu, Danzhen.
2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009|Int. Conf. Artif. Intell. Comput. Intell., AICI. Vol. 2 2009. p. 511-514.

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

TY - GEN

T1 - Day-ahead electricity prices forecasting using artificial neural networks

AU - Tang, Qi

AU - Gu, Danzhen

N1 - Artificial neural network;Bidding strategy;Deregulated electricity market;Electricity prices;Electricity prices forecasting;Generating companies;Learning mechanism;Market Clearing Price;Market clearing prices;Nonlinear problems;Training methods;

PY - 2009

Y1 - 2009

N2 - This paper presents specifically how artificial neural network (ANN) is applied to forecast day-ahead electricity price in the deregulated electricity market. Market clearing price (MCP) forecasting has been more and more significant in new restructured market because both generating companies and consumers rely on it to prepare their bidding strategies and expect to maximize respective benefits with low risks. But the prediction of MCP is complex because various uncertainties interact in an intricate way. Hence, ANN is proposed to solve high complicated nonlinear problems like MCP forecasting due to its powerful capability of learning mechanism if enough data for training. The paper focuses on influences brought by different input architectures with different training methods and proposes a basic optimal ANN architecture for day-ahead MCP forecasting. © 2009 IEEE.

AB - This paper presents specifically how artificial neural network (ANN) is applied to forecast day-ahead electricity price in the deregulated electricity market. Market clearing price (MCP) forecasting has been more and more significant in new restructured market because both generating companies and consumers rely on it to prepare their bidding strategies and expect to maximize respective benefits with low risks. But the prediction of MCP is complex because various uncertainties interact in an intricate way. Hence, ANN is proposed to solve high complicated nonlinear problems like MCP forecasting due to its powerful capability of learning mechanism if enough data for training. The paper focuses on influences brought by different input architectures with different training methods and proposes a basic optimal ANN architecture for day-ahead MCP forecasting. © 2009 IEEE.

KW - Artificial neural network

KW - Forecasting

KW - Market clearing price

UR - https://www.scopus.com/pages/publications/77949302249

M3 - Conference contribution

SN - 9780769538167

VL - 2

SP - 511

EP - 514

BT - 2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009|Int. Conf. Artif. Intell. Comput. Intell., AICI

T2 - 2009 International Conference on Artificial Intelligence and Computational Intelligence, AICI 2009

Y2 - 1 July 2009

ER -

Day-ahead electricity prices forecasting using artificial neural networks

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this