Estimation of natural gas consumption using artificial neural network: A case study in Ankara

Correct estimation of natural consumption is crucial for both gas distributors and consumers. By means of an accurate estimation, potential issues regarding natural gas distribution systems may be reduced, distribution limits may be planned properly. In order to perpetuate gas procurement, accurate estimation model is extremely important for the countries which has not storage system like Turkey and dependent on foreign sources reaching to high values like 95% (39 billion m(3)/year). In this paper, an ampiric relation has been developed dependent on basic consumption indicators and meteorological data by making use of Artificial Neural Networks (ANN) which is used broadly in various disciplines and providing good estimation results in nonlinear multivariable models. With the proposed system, monthly natural gas consumption estimation study is implemented for Ankara which is the capital city and is one of the first-time users of natural gas in dwelling houses (share in total consumption 21.3%, 9.8% in terms of customer number). R-2=1 statistical error value is found for analytical expression of provided estimation model.