Artificial bee colony algorithm to forecast natural gas consumption of Turkey

Abstract

Natural gas is one of the most important energy sources in the world because it is cheaper and cleaner than most of its alternatives. Therefore, natural gas is used for residential and industrial purposes such as heating, electricity generation and manufacturing. Although Turkey has its own resources to meet the demand for natural gas, most of the needed natural gas amount is imported from other countries. The produced or imported natural gas has been used for heating, electricity generation and manufacturing in Turkey. Therefore approximate forecasting with a less statistical error of vital energy sources such as natural gas is extremely significant for countries having an imbalance between export and import values. The main purpose of this study is to determine the best coefficients with less statistical error for the forecasting equation for natural gas demand of Turkey. This paper suggests an Artificial Bee Colony (ABC) algorithm for forecasting natural gas consumption in billion cubic meters (bcm) for the years between 2018 and 2030 by using some historical indicators between 1998 and 2017. ABC algorithm is firstly used for determining coefficients of a forecasting equation for natural gas demand of Turkey. ABC algorithm was originally designed for continuous optimization problems and with its simple structure and applicability can be useful for decision or policymakers. Therefore such a strong metaheuristic algorithm in the literature should be used for a vital problem of the countries having current account deficit like Turkey. Annual gross domestic product at chained volume index, population, export and import of Turkey are selected as historical indicators and independent variables for forecasting in this study. Furthermore, a linear regression model is also created to compare the proposed forecasting model. Both models are used to forecast Turkey future natural gas consumption for two different scenarios. The proposed algorithm outperforms the linear regression model in view of total absolute relative errors from previous natural gas consumptions of Turkey.

A Novel Grey Seasonal Model for Natural Gas Production Forecasting

“…Based on the Ant Lion Optimization algorithm (ALO), an adaptive grey prediction system with nonlinear optimization initial value was proposed [14]. An artificial bee colony (ABC) algorithm was designed to solve the continuous optimization problem, and the natural gas consumption of Turkey from 2018 to 2030 is predicted [15]. To forecast the short-term consumption of natural gas, a new hybrid prediction model is established by combining Volterra adaptive filtering with an improved whale optimization algorithm [16].…”

“…The detailed calculation process is as follows. 6,7,8,10,11,12,14,15,16,18,19,20) can be obtained in the same way.…”

Electric power load forecasting on a 33/11 kV substation using artificial neural networks

“…Performance of the neural network during training and testing is observed based on metric like mean square error (MSE) and mean absolute percentage error (MAPE) [24] as shown in Eqs. (1) and (2) respectively.…”

Güneydoğu Avrupa Bölgesi̇ Doğal Gaz Pi̇yasasinin Ukrayna-Rusya Savaşi Sonrasi Ekopoli̇ti̇ği̇

“…) ise kısa vadeli enerji talebinin doğru tahmini için Yapay Sinir Ağları ve zaman serileri yöntemlerini kullanmıştır.Eren ve Kaçtıoğlu (2017), TÜİK'de yer alan Türkiye'nin yıllık doğal gaz tüketim ve nüfus verilerini inceleyerek, Gri model (1,1) modeli aracılığı ile 2017 -2030 dönemi için nüfus ve doğal gaz tüketim tahminleri elde etmişlerdir Arık (2019),. Türkiye'nin doğal gaz talebi tahmin denklemi için daha az istatistiksel hata ile en iyi katsayıları belirlemeye çalışmıştır.Canyurt ve Öztürk (2008) ile Ervural vd.…”