Journals Information
Universal Journal of Engineering Science(CEASE PUBLICATION) Vol. 2(4), pp. 73 - 76
DOI: 10.13189/ujes.2014.020402
Reprint (PDF) (243Kb)
Forecasting Gas Production Rate in Underground Gas Reservoirs Using Artificial Neural Networks
Morteza Bagherpour 1,*, MH.Bagherpour 2, K. Roodani 2
1 Department of industrial Engineering, Iran University of Science and Technology, Iran
2 JondiShapour Company, Shiraz, Iran
ABSTRACT
Underground Gas Reservoir (UGS) is one of the important types of reservoirs which after depletion, a significant amount of gas will inject into the reservoir. Then, in winter and in emergency cases, the stored gas will transfer to consumption centers. Here, forecasting of gas production rate will play an important role to discover when the reservoir will be depleted and, how much stored gas amount will be then injected. The forecasting procedure of UGS highly depends on affecting factors such as pressure, gas in place, bulk volume and etc. Due to complexity and non-linearity manner of input and output variables, an intelligent data driven modeling procedure is useful to apply. In this paper, therefore, a Multi Layer Perceptron (MLP) neural network is applied to predict the production rate. Also, a time series model is conducted to predict time based variables. The results show a 99 percent of accuracy which demonstrate superiority of the proposed approach over existing traditional models.
KEYWORDS
UGS, Gas Production Rate, Forecasting, Neural Networks, MLP, Time Series Analysis
Cite This Paper in IEEE or APA Citation Styles
(a). IEEE Format:
[1] Morteza Bagherpour , MH.Bagherpour , K. Roodani , "Forecasting Gas Production Rate in Underground Gas Reservoirs Using Artificial Neural Networks," Universal Journal of Engineering Science(CEASE PUBLICATION), Vol. 2, No. 4, pp. 73 - 76, 2014. DOI: 10.13189/ujes.2014.020402.
(b). APA Format:
Morteza Bagherpour , MH.Bagherpour , K. Roodani (2014). Forecasting Gas Production Rate in Underground Gas Reservoirs Using Artificial Neural Networks. Universal Journal of Engineering Science(CEASE PUBLICATION), 2(4), 73 - 76. DOI: 10.13189/ujes.2014.020402.