Abstract:Based on the characteristics of natural gas demand trend, this paper proposed ARIMA model which can predict China's natural gas demand as an effective tool. Compared with the RBF neural network model and combined model, empirical results show that the accuracy and stability of the ARIMA model is best.
“…Authors who used neural networks were [24], [27], [12], [29], [34], [14], [43], [15], [40], [39], and [32]. Viet & Mandziuk [24] presented several neural and fuzzy neural approaches.…”
Section: Overview Of Prediction Methodsmentioning
confidence: 99%
“…Olgun et al [14] compared neural networks with support vector machines and they concluded that SVM had less statistical error. Feng et al [15] developed three different kinds of model -ARIMA model, neural network model and combined model. Neural network model (radial basis function) achieved the MAPE of 5.78%.…”
Section: Overview Of Prediction Methodsmentioning
confidence: 99%
“…Traditional grey prediction model resulted with the relative error of 10.5%, while the same error based on the optimized model was smaller (8.16%). As mentioned earlier, Feng et al [15] developed three different kinds of model -ARIMA model, neural network model and combined model. The MAPE of the combined (hybrid) model was 4.52%.…”
Due to its many advantages, demand for natural gas has increased considerably and many models for predicting natural gas consumption are developed. The aim of this paper is to present an overview and systematic analysis of the latest research papers that deal with predictions of natural gas consumption for residential and commercial use from the year 2002 to 2017. Literature overview analysis was conducted using the two most relevant scientific databases Web of Science Core Collection and Scopus. The results indicate neural networks as the most common method used for predictions of natural gas consumption, while most accurate methods are genetic algorithms, support vector machines and ANFIS. Most used input variables are past natural gas consumption data and weather data, and prediction is most commonly made on daily and annual level on a country area level. Limitations of the research raise from relatively small number of analyzed papers but still research could be used for significant improving of prediction models for natural gas consumption.
“…Authors who used neural networks were [24], [27], [12], [29], [34], [14], [43], [15], [40], [39], and [32]. Viet & Mandziuk [24] presented several neural and fuzzy neural approaches.…”
Section: Overview Of Prediction Methodsmentioning
confidence: 99%
“…Olgun et al [14] compared neural networks with support vector machines and they concluded that SVM had less statistical error. Feng et al [15] developed three different kinds of model -ARIMA model, neural network model and combined model. Neural network model (radial basis function) achieved the MAPE of 5.78%.…”
Section: Overview Of Prediction Methodsmentioning
confidence: 99%
“…Traditional grey prediction model resulted with the relative error of 10.5%, while the same error based on the optimized model was smaller (8.16%). As mentioned earlier, Feng et al [15] developed three different kinds of model -ARIMA model, neural network model and combined model. The MAPE of the combined (hybrid) model was 4.52%.…”
Due to its many advantages, demand for natural gas has increased considerably and many models for predicting natural gas consumption are developed. The aim of this paper is to present an overview and systematic analysis of the latest research papers that deal with predictions of natural gas consumption for residential and commercial use from the year 2002 to 2017. Literature overview analysis was conducted using the two most relevant scientific databases Web of Science Core Collection and Scopus. The results indicate neural networks as the most common method used for predictions of natural gas consumption, while most accurate methods are genetic algorithms, support vector machines and ANFIS. Most used input variables are past natural gas consumption data and weather data, and prediction is most commonly made on daily and annual level on a country area level. Limitations of the research raise from relatively small number of analyzed papers but still research could be used for significant improving of prediction models for natural gas consumption.
In recent years, natural gas utilisation has seen a considerable increase because, it presents an alternative energy source that is reliable, economical and environmentally friendly for consumers. In Ghana, natural gas consumption has over the years increased due to mainly the rise in industrial and residential demands. Accurate prediction of natural gas consumption will provide stakeholders with vital information needed for planning and making informed policy decisions. This paper explores the Autoregressive Integrated Moving Average (ARIMA) and Seasonal Autoregressive Integrated Moving Average (SARIMA) to predict Ghana's daily natural gas consumption. The data employed for the study is daily natural gas consumption in Ghana from 2020 to 2022. The results show that both ARIMA and SARIMA models can predict the consumption of natural gas in Ghana with a good degree of accuracy. The SARIMA model slightly outperforms the ARIMA model, with a Root Mean Square Error (RMSE) of 22.25 and a Mean Absolute Percentage Error (MAPE) of 6.96%, compared to an RMSE of 23.27 and a MAPE of 7.29% for the ARIMA model. The model forecast suggests a steady natural gas consumption in Ghana but with some intermittent fluctuations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.