Forecasting of the daily demand for natural gas in rural households using the methods of artificial intelligence. Part I. Forecasting using artificial neural networks

K. Nęcka; M. Trojanowska; J. Małopolski

2/2015 vol. 60

ORIGINAL PAPER

K. Nęcka ¹

M. Trojanowska ¹

J. Małopolski ¹

More details

Hide details

University of Agriculture in Krakow, Department of Power Engineering and Agricultural Processes Automation, ul. Balicka 116B, 30-149 Kraków, Poland

Journal of Research and Applications in Agricultural Engineering 2015;60(2):62-64

Article (PDF)

References (14)

KEYWORDS

natural gas

short-term forecasts

artificial neural networks

ABSTRACT

The paper determines daily forecast demands for natural gas using artificial neural networks (MLPs). The influence of net-work structure, the type of activation function and the training process used on the quality of prediction were studied. It was found that the quality of forecasts was highly influenced by the network training algorithm. The smallest errors of the ex-pired forecasts (MAPE 5-6%) were obtained using the BFGS algorithm.

REFERENCES (14)

Azari A.; Shariaty-Niassar M., Alborzi M.: Short-term and medium-term gas demand load forecasting by neural networks. Iranian Journal of Chemistry and Chemical Engineering, 2012, vol. 31(4), 77-84.

Google Scholar

Chen Q., She Y., Xu X.: Combination model for short-term load forecasting. The Open Automation and Control Systems Journal, 2013, vol. 5, 124-132.

Google Scholar

Dittman P., Szabela-Pasierbińska E.: Short-term sales fore-casts in management of natural gas distributing works. Management, 2007, vol. 11(1), 147-154.

Google Scholar

Dyrektywa Parlamentu Europejskiego i Rady 2009/73/WE z dnia 13 lipca 2009 r. dotycząca wspólnych zasad rynku wewnętrznego gazu ziemnego. Dziennik Urzędowy Unii Euro-pejskiej.14.8.2009. L211/94-L211/136.

Google Scholar

Kelner J.M.: Prognozowanie krótkoterminowe poborów gazu z sieci przesyłowych metodą sztucznych sieci neuronowych. Gaz, Woda i Technika Sanitarna, 2003, 6, 196-204.

Google Scholar

Kizilaslan R., Karlik B.: Comparison neural networks models for short term forecasting of natural gas consumption in Is-tanbul. In: Applications of digital information and web technologies, ICADIWT, 2008, 4-6 August 2008, 448-453.

Google Scholar

Nęcka K., Trojanowska M.: Short-term forecasting of natural gas demand by rural consumers using regression models. TEKA, 2014.

Google Scholar

Potocnik P., Govekar E., Grabec I.: Short-term natural gas consumption forecasting. In: Proceedings of the 16th IASTED International Conference on Applied Simulation and Modelling - ASM, 2007, 353-357.

Google Scholar

Soldo B.: Forecasting natural gas consumption. Applied Energy, 2012, vol. 92, 26-37.

Google Scholar

10.

Trojanowska M., Lipczyńska D.: Analysing the variation in demands of rural households for natural gas. TEKA, 2014.

Google Scholar

11.

Wdrożenie kodeksu bilansowania - szanse i wyzwania dla rynku gazu w obszarze dystrybucji. Polska Spółka Gazownictwa: TGE Warszawa, 2014.

Google Scholar

12.

Yu F., Xu X.: A short-term load forecasting model of natural gas based on optimized genetic algorithm and improved BP neural network. Applied Energy, 2014, vol. 134, 102-113.

Google Scholar

13.

Zeliaś A., Pawełek B., Wanat S.: Prognozowanie ekonomiczne. Warszawa: PWN, 2004. ISBN 83-01-14043-7.

Google Scholar

14.

Zhou H., Su G., Li G.: Forecasting daily gas load with OIHF-Elman Neural Network. Procedia Computer Science, 2011, vol. 5, 754-758.

Google Scholar

Submit your paper

Instructions for Authors

eISSN:	2719-423X
ISSN:	1642-686X