Using an artificial neural network model for natural gas compositions forecasting

Jolanta Szoplik and Paulina Muchel

Energy, 2023, vol. 263, issue PD

Abstract: The paper presents the results of natural gas composition forecasting obtained using the MLP model of artificial neural network. The training of MLP model was performed on the basis of 8760 real data describing the percentage shares of the five main components of natural gas in a selected town on the territory of Poland. The model includes calendar factors (month, day of month, day of week, hour of day) and weather factors (ambient temperature), which indirectly affect the composition of natural gas in a given point of the gas network and were selected after a detailed statistical analysis of the variability of the natural gas composition in time. Based on the value of the correlation coefficient for the test set and the MAPE forecast errors calculated on the basis of the actual and the forecast data, the best quality MLP 18-65-5 network was experimentally selected. Natural gas composition forecasts were made using this model for input data characterizing the next calendar year and the average MAPE forecast error = 3.356% was calculated. Risk analysis, in turn, was used to estimate the probability of obtaining a forecast with a MAPE error greater than the mean error.

Keywords: Natural gas composition variability; Natural gas compositions forecasting; Artificial neural network model; Risk analysis (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544222028870
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:263:y:2023:i:pd:s0360544222028870

DOI: 10.1016/j.energy.2022.126001

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().