Page - 175 - in Short-Term Load Forecasting by Artificial Intelligent Technologies

Energies2018,11, 2038 agreedwith theretailer (whichrefers to thevalueof theenergyconsumed).Note that theconcepts corresponding toaccess fees (power andenergy terms) and taxes are independent of themodeof supply, so theydonot change for aDirectConsumer. Therefore, the calculationof the cost of the referenced energy for theDMCand its comparisonwith the retailer cost, is themain concern for this study. Recall that,under theassumptionsof this study, thehourlyenergycostasaDMCisgivenbythe sumof fourcomponents: thecost in theDailyMarket (DMcost), theadjustmentservices (AScost), themeasureddeviations (MDcost), and thecapacitypayments (CPcost). Table9 shows thevalue of eachcomponentwhen thecostof theenergyasaDMCisevaluated. In this section, 48-h-ahead predictionsobtainedwith theXGBoostmethod(eta=0.02,nrounds=3700)wereused,althoughany of theotherensemblemethodswould leadtosimilar results. It isworth tomention that thecostof deviations isquite limiteddueto theaccuracyof the loadforecastingmethod. Table9.MonthlycostactingasaDirectMarketConsumers (DMC)anditscomponents. Month DMCost (in€) ASCost (in€) MDCost (in€) CPCost (in€) DMCCost (in€) January 5478 685 91 313 6567 February 4492 815 56 409 5772 March 3644 763 45 99 4551 April 2980 649 70 105 3804 May 3976 801 43 127 4948 June 6692 682 42 336 7752 July 7151 610 28 524 8313 August 4450 430 56 0 4936 September 8013 724 57 195 8989 October 8289 708 48 130 9176 November 7960 474 91 140 8665 December 8727 492 46 285 9575 Total2016 71,853 (86.52%) 7834 (9.44%) 697 (0.84%) 2664 (3.2%) 83,048 Table10showstheelectricityconsumption (inkWh)of thecampusuniversity in2016andthe costof the referencedenergy (consumption) in fourcases: the real costpaid to the retailer, thecost usingtheAverageFinalPrice (AFP),actingasaDMC,andwhatwecall thepessimistprice (aDirect Consumerwithall thedeviationsagainst thesystem).Accordingto theresults, it canbeestablished thatDMCmodalitywouldhaveproducedsavingsofaround11%intheenergytermof the invoice whencomparedto theretailprice.Notealso that thecostusingtheAFPdoesnotcoincidewith the costof theDMCbecause thecostdue todeviationsandthecapacitypaymentscomponentsdependon theconsumer.Ontheotherhand, theresults showthat, even in thepessimistic case (alldeviationsof thepredictionsagainst thesystem), theDMCtypeofsupply isworthyagainst theretailer. It is important tohighlight that the economicbenefits of theDMCtypeof supplydependon twomainaspects: themagnitudeof thedeviationsand thedirectionof thedeviations (towardsor against thesystem). Thefirstaspect (magnitudeof thedeviations) isdeterminedbytheaccuracyof the forecastingmethod.However, thesecondaspect (directionof thedeviations) isoutofourcontroland itdependsonthewholeElectricSystem. Inparticular, someworse forecastingmethodscould leadto greaterbenefits thanmoreaccuracymethods,butonlybychanceandassumingthat the forecasting valuesaregoodenough(moderatedeviations). Therefore, the loadforecastingmethodis important to someextent,butobviously lowerdeviationsarepreferable togreaterdeviations. 175