{"title":"Electricity Load Modeling: An Application to Italian Market","authors":"Giovanni Masala, Stefania Marica","volume":93,"journal":"International Journal of Economics and Management Engineering","pagesStart":3149,"pagesEnd":3159,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/10003944","abstract":"Forecasting electricity load plays a crucial role regards
\r\ndecision making and planning for economical purposes. Besides, in
\r\nthe light of the recent privatization and deregulation of the power
\r\nindustry, the forecasting of future electricity load turned out to be a
\r\nvery challenging problem. Empirical data about electricity load
\r\nhighlights a clear seasonal behavior (higher load during the winter
\r\nseason), which is partly due to climatic effects. We also emphasize
\r\nthe presence of load periodicity at a weekly basis (electricity load is
\r\nusually lower on weekends or holidays) and at daily basis (electricity
\r\nload is clearly influenced by the hour). Finally, a long-term trend may
\r\ndepend on the general economic situation (for example, industrial
\r\nproduction affects electricity load). All these features must be
\r\ncaptured by the model.
\r\nThe purpose of this paper is then to build an hourly electricity load
\r\nmodel. The deterministic component of the model requires non-linear
\r\nregression and Fourier series while we will investigate the stochastic
\r\ncomponent through econometrical tools.
\r\nThe calibration of the parameters’ model will be performed by
\r\nusing data coming from the Italian market in a 6 year period (2007-
\r\n2012). Then, we will perform a Monte Carlo simulation in order to
\r\ncompare the simulated data respect to the real data (both in-sample
\r\nand out-of-sample inspection). The reliability of the model will be
\r\ndeduced thanks to standard tests which highlight a good fitting of the
\r\nsimulated values.","references":"[1] Afshar K. and Bigdeli N. (2011) \"Data analysis and short term load\r\nforecasting in Iran electricity market using singular spectral analysis\r\n(SSA)\", Energy, Vol. 36, pp. 2620-2627.\r\n[2] Alter N. and Syed H. S. (2011) \u201cAn empirical analysis of electricity\r\ndemand in Pakistan\u201d, International Journal of Energy Economics and\r\nPolicy, Vol. 1, No. 4, pp. 116-139.\r\n[3] Andersson G., Cornel J., Hezog F., Hildmann, M. and Stokic, D. (2011)\r\n\u201cRobust calculation and parameter estimation of the Hourly Price\r\nForward Curve\u201d, 17th Power Systems Computation Conference, August\r\n22-26, 2011, Stockholm, Sweden.\r\n[4] Andersson G., He, Y., Hildmann M. and Sotiropoulos E. (2013)\r\n\u201cModeling of Electricity Load for Forward Contract Pricing\u201d, Institute\r\nof Electrical and Electronics Engineers, IEEE, July 21, 2013.\r\n[5] Bianco V., Manca O. and Nardini S. (2009) \u201cElectricity consumption\r\nforecasting in Italy using Linear Regression Models\u201d, Energy, Vol. 34,\r\npp. 1413-1421.\r\n[6] Bilgili M., Sahin B., Yasar A. and Simsek E. (2012) \u201cElectric energy\r\ndemands of Turkey in residential and industrial sectors\u201d, Renewable and\r\nSustainable Energy Reviews, Vol. 16, pp. 404-414.\r\n[7] Bl\u00e0zquez L., Nina B. and Filippini M. (2012) \u201cResidential electricity\r\ndemand for Spain: new empirical evidence using aggregated data\u201d,\r\nCentre for Energy Policy and Economics, Swiss Federal Institutes of\r\nTechnology, CEPE working paper No. 82.\r\n[8] Bruhns A., Deurveilher G., and Roy J. S. \u201cA non-linear regression\r\nmodel for mid-term load forecasting and improvements in seasonality\u201d,\r\n15th PSCC, Liege, 22-26 August 2005.\r\n[9] Chujai P., Kerdprasop N. and Kerdprasop K. \u201cTime series analysis of\r\nhousehold electric consumption with ARIMA and ARMA models\u201d,\r\nIMECS 2013, March 13-15, 2013, Hong Kong.\r\n[10] Collet J., Dessertaine A., Dordonnat V., Koopman S. J. And Ooms M.,\r\n(2008) \u201cJournal of Forecasting\u201d, Vol. 24, pp. 566-587.\r\n[11] Deihimi A., Orang O. and Showkati H. (2012) \"Short-term electric load\r\nand temperature forecasting using wavelet echo state networks with\r\nneural reconstruction\" , Energy, Vol. 57, pp. 382-401.\r\n[12] Fan S. and Hyndman R. J. (2012) \u201cShort-term load forecasting based on\r\na semi-parametric additive model\u201d, IEEE Transactions on Power\r\nSystems, Vol. 27, No. 1, pp. 134-141.\r\n[13] Filik U. B., Gerek \u00d6., N. and Kurban M. (2011) \u201cA novel modeling\r\napproach for hourly forecasting of long-term electric energy demand\u201d,\r\nEnergy Conversion and Management, Vol. 52, pp. 199-211.\r\n[14] Gonz\u00e0les-Romera E., Jaramillo-Mor\u00e0n M. A. and Carmona-Fern\u00e0ndez\r\nD. (2008) \"Monthly electric energy demand forecasting with neutral\r\nnetworks and Fourier series\", Energy Conversion and Management, Vol.\r\n49, pp. 3135-3142.\r\n[15] Hong W-C (2011) \"Electric load forecasting by seasonal recurrent SVR\r\n(support vector regression) with chaotic artificial bee colony algorithm\",\r\nEnergy, Vol. 36, pp. 5568-5578.\r\n[16] Kavousian A., Rajagopal R. and Fischer M. (2013) \u201cDeterminants of\r\nresidential electricity consumption: using smart meter data to examine\r\nthe effect of climate, building characteristics, appliance stock, and\r\noccupants\u2019 behavior\u201d, Energy, Vol. 55, pp. 184-194.\r\n[17] Makridakis S., Wheelwright S. C., and Hyndman R. J. (1998)\r\n\u201cForecasting: methods and applications\u201d New York: John Wiley &\r\nSons.\r\n[18] Migon H. S., and Alves L. C., (2013) \u201cMultivariate dynamic regression:\r\nmodeling and forecasting for intraday electricity load\u201d, Applied\r\nStochastic Models in Business and Industry, Vol. 29, pp. 579-598.\r\n[19] Moral-Carcedo J. and Vic\u00e9ns-Otero J. (2005) \u201cModelling the non-linear\r\nresponse of Spanish electricity demand to temperature variations\u201d,\r\nEnergy Economics, Vol. 27, Issue 3, pp. 477-494.\r\n[20] Nagi J., Yap K.S., Tiong S.K., Ahmed, S.K. \u201cElectrical power load\r\nforecasting using hybrid self-organizing maps and support vector\r\nmachines\u201d, The 2nd International Power Engineering and Optimization\r\nConference (PEOCO 2008), Shah Alam, Selangor, Malaysia, 4-5 June\r\n2008.\r\n[21] Pardo, A. Meneu V., and Valor E. (2002) \u201cTemperatures and seasonality\r\ninfluences on Spanish electricity load\u201d, Energy Economics, Vol. 24, pp.\r\n55-70.\r\n[22] Pielow A, Sioshansi R. and Roberts M.C. (2012) \u201cModeling short-run\r\nelectricity demand with long-term growth rates and consumer price\r\nelasticity in commercial and industrial sectors\u201d, Energy, Vol. 46,\r\npp.533-540.\r\n[23] Saab S., Badr E., and Nasr G. (2001) \u201cUnivariate modeling and\r\nforecasting of energy consumption: the case of electricity in Lebanon\u201d,\r\nEnergy, Vol. 26, pp. 1-14.\r\n[24] Soares L. J. and Medeiros M. (2008) \u201cModeling and forecasting shortterm\r\nelectricity load: A comparison of methods with an application to\r\nBrazilian data\u201d, International Journal of Forecasting, Vol. 24, pp. 630-\r\n644.\r\n[25] Wang C-H., Grozev G. and Seo S. (2012) \"Decomposition and statistical\r\nanalysis for regional electricity demand forecasting\", Energy, Vol. 41,\r\npp. 313-325.\r\n[26] Weron R., (2006), \u201cModeling and forecasting electricity loads and\r\nprices. A statistical approach\u201d, John Wiley & Sons Ltd.\r\n[27] http:\/\/www.mercatoelettrico.org\/En\/Default.aspx\r\n[28] http:\/\/www.istat.it\/en","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 93, 2014"}