Natural Gas Consumption Forecasting Based on the Variability of External Meteorological Factors Using Machine Learning Algorithms

Natural gas consumption depends on many factors. Some of them, such as weather conditions or historical demand, can be accurately measured. The authors, based on the collected data, performed the modeling of temporary and future natural gas consumption by municipal consumers in one of the medium-sized cities in Poland. For this purpose, the machine learning algorithms, neural networks and two regression algorithms, MLR and Random Forest were used. Several variants of forecasting the demand for natural gas, with different lengths of the forecast horizon are presented and compared in this research. The results obtained using the MLR, Random Forest, and DNN algorithms show that for the tested input data, the best algorithm for predicting the demand for natural gas is RF. The differences in accuracy of prediction between algorithms were not significant. The research shows the differences in the impact of factors that create the demand for natural gas, as well as the accuracy of the prediction for each algorithm used, for each time horizon.

A Historical and Econometric Analysis of Energy Consumption and Industrial Output in Pakistan (1990-2019)

This paper examines the nexus of disaggregated energy consumption and industrial output in Pakistan. The annual time series data over the period 1990-2019 has been taken for current research. ARDL technique has been employed for empirical analysis. The results show that oil consumption, electricity consumption and gas consumption are positively and significantly connected with the industrial output in long run. Similarly, trade openness, labour and capital also have the same association with the industrial output and have significant outcomes in the long run. The results of Granger causality show that there exists a unidirectional causality from electricity consumption to industrial output. The study concludes that oil, gas and electricity are contributing a large share in industrial growth so that it would be made an effort to install the plants relevant with these energy sources to meet the affordable demand in the industry sector.

Scenario Prediction of China’s Natural Gas Consumption and Carbon Emissions in the Next Ten Years

Based on Johansen Cointegration Test, this paper sheds light on the long-run equilibrium relationship between natural gas consumption, gas production, and GDP in China. Three different natural gas demand scenarios of low, medium and high rates in the next ten years are considered, and a Neural Network Autoregression Model is used to predict the future carbon dioxide emission. We conclude: (1) In all three scenarios, the growth rates of natural gas consumption are all higher than those of natural gas production, while the gap between demand and domestic supply will gradually turn broader and China will largely rely on imports ; (2) In the scenario of low-rate economic growth, natural gas consumption will grow slowly, and it will be difficult to realize the carbon emission reduction targets by 2030 due to low-rate substitution of natural gas for coal; (3) If medium-rate to high-rate economic growth sustains, coupled with rapid increase in natural gas consumption and production, China’s Carbon Emission Reduction Targets for 2030 can be achieved with high-rate substitution of natural gas for coal.

Forecasting of Natural Gas Consumption in Poland Based on ARIMA-LSTM Hybrid Model

Natural gas is one of the main energy sources in Poland and accounts for about 15% of the primary energy consumed in the country. Poland covers only 1/5 of its demand from domestic deposits. The rest is imported from Russia, Germany, Norway, the Czech Republic, Ukraine, and Central Asia. An important issue concerning the market of energy resources is the question of the direct impact of the prices of energy resources on the income of exporting and importing countries. It should also be remembered that unexpected and large fluctuations are detrimental to both exporters and importers of primary fuels. The article analyzes natural gas deposits in Poland, raw material trade and proposes a model for forecasting the volume of its consumption, which takes into account historical consumption, prices of energy resources and assumptions of Poland’s energy policy until 2040. A hybrid model was built by combining ARIMA with LSTM artificial neural networks. The validity of the constructed model was assessed using ME, MAE, RMSE and MSE. The average percentage error is 2%, which means that the model largely represents the real gas consumption course. The obtained forecasts indicate an increase in consumption by 2040.