A comparison of models for forecasting the residential natural gas demand of an urban area

Abstract Forecasting the residential natural gas demand for large groups of buildings is extremely important for efficient logistics in the energy sector. In this paper different forecast models for residential natural gas demand of an urban area were implemented and compared. The models forecast gas demand with hourly resolution up to 60 h into the future. The model forecasts are based on past temperatures, forecasted temperatures and time variables, which include markers for holidays and other occasional events. The models were trained and tested on gas-consumption data gathered in the city of Ljubljana, Slovenia. Machine-learning models were considered, such as linear regression, kernel machine and artificial neural network. Additionally, empirical models were developed based on data analysis. Two most accurate models were found to be recurrent neural network and linear regression model. In realistic setting such trained models can be used in conjunction with a weather-forecasting service to generate forecasts for future gas demand.

[1]  Shahaboddin Shamshirband,et al.  Intelligent forecasting of residential heating demand for the District Heating System based on the monthly overall natural gas consumption , 2015 .

[2]  Friederich Kupzog,et al.  Building power demand forecasting using K-nearest neighbors model - initial approach , 2016, 2016 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC).

[3]  Selim Zaim,et al.  Forecasting Electricity Consumption with Neural Networks and Support Vector Regression , 2012 .

[4]  Luca A. Tagliafico,et al.  Heating and cooling building energy demand evaluation; a simplified model and a modified degree days approach , 2014 .

[5]  Ahmet Serdar Yilmaz,et al.  Long Term Energy Consumption Forecasting Using Genetic Programming , 2008 .

[6]  Edvard Govekar,et al.  Applied Forecasting of Short-Term Natural Gas Consumption , 2011 .

[7]  Božidar Soldo,et al.  Forecasting natural gas consumption , 2012 .

[8]  Marco De Nadai,et al.  Short-term anomaly detection in gas consumption through ARIMA and Artificial Neural Network forecast , 2015, 2015 IEEE Workshop on Environmental, Energy, and Structural Monitoring Systems (EESMS) Proceedings.

[9]  Shahaboddin Shamshirband,et al.  Prediction of heat load in district heating systems by Support Vector Machine with Firefly searching algorithm , 2016 .

[10]  Abdul Hanan Abdullah,et al.  Heat load prediction in district heating systems with adaptive neuro-fuzzy method , 2015 .

[11]  Soteris A. Kalogirou,et al.  Artificial neural networks for the prediction of the energy consumption of a passive solar building , 2000 .

[12]  Masatoshi Sakawa,et al.  Heat load prediction through recurrent neural network in district heating and cooling systems , 2008, 2008 IEEE International Conference on Systems, Man and Cybernetics.

[13]  A Aleksandra Sretenovic,et al.  Multistage ensemble of feedforward neural networks for prediction of heating energy consumption , 2016 .

[14]  Shahaboddin Shamshirband,et al.  Forecasting of consumers heat load in district heating systems using the support vector machine with a discrete wavelet transform algorithm , 2015 .

[15]  Liljana Ferbar Tratar,et al.  The comparison of Holt–Winters method and Multiple regression method: A case study , 2016 .

[16]  Frédéric Magoulès,et al.  A review on the prediction of building energy consumption , 2012 .

[17]  David Hsu,et al.  Comparison of integrated clustering methods for accurate and stable prediction of building energy consumption data , 2015 .

[18]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[19]  S. Saeedeh Sadegh,et al.  Forecasting energy consumption using ensemble ARIMA-ANFIS hybrid algorithm , 2016 .

[20]  Corrado Schenone,et al.  Model for forecasting residential heat demand based on natural gas consumption and energy performance indicators , 2016 .

[21]  Peter Tzscheutschler,et al.  Short-term smart learning electrical load prediction algorithm for home energy management systems , 2015 .

[22]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[23]  Marek Brabec,et al.  A statistical model for the estimation of natural gas consumption , 2008 .

[24]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[25]  Razvan Pascanu,et al.  On the Number of Linear Regions of Deep Neural Networks , 2014, NIPS.

[26]  Tomislav Šarić,et al.  Comparison of static and adaptive models for short-term residential natural gas forecasting in Croatia , 2014 .

[27]  P. K. Adom,et al.  Conditional dynamic forecast of electrical energy consumption requirements in Ghana by 2020: A comparison of ARDL and PAM , 2012 .

[28]  Kelvin K. W. Yau,et al.  Predicting electricity energy consumption: A comparison of regression analysis, decision tree and neural networks , 2007 .

[29]  Shengwei Wang,et al.  Development of prediction models for next-day building energy consumption and peak power demand using data mining techniques , 2014 .

[30]  Lin Jiang,et al.  Short-term natural gas demand prediction based on support vector regression with false neighbours filtered , 2015 .

[31]  Sousso Kelouwani,et al.  Estimation of temperature correlation with household electricity demand for forecasting application , 2016, IECON 2016 - 42nd Annual Conference of the IEEE Industrial Electronics Society.

[32]  Tao Hong,et al.  Probabilistic electric load forecasting: A tutorial review , 2016 .

[33]  Richard F. Gunst,et al.  Applied Regression Analysis , 1999, Technometrics.

[34]  M. Wand,et al.  Multivariate plug-in bandwidth selection , 1994 .

[35]  Ian T. Nabney,et al.  Short-term electricity demand and gas price forecasts using wavelet transforms and adaptive models , 2010 .

[36]  Yan Lu,et al.  Modeling and forecasting of cooling and electricity load demand , 2014 .

[37]  Flora D. Salim,et al.  Multivariate electricity consumption prediction with Extreme Learning Machine , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).