Page - 148 - in Short-Term Load Forecasting by Artificial Intelligent Technologies

Energies2018,11, 2080 -20 0 20 40 2 2.5 3 3.5 4 4.5 x 10 4 Temp. at Madrid (ºC) 0 20 40 2 2.5 3 3.5 4 4.5 x 10 4 Temp. at Barcelona (ºC) Linearization of Load Against Temperature 0 20 40 2 2.5 3 3.5 4 4.5 x 10 4 Temp at Sevilla (ºC) Figure6.Scatterplotofnational loadandits linearizationagainst temperatureatMadrid,Barcelona andSevilla. 3.3. Calendar ThetypeofdayisdeterminedbytheofficialnationalcalendarpublishedintheOfficialGazette[46]. Thedaysareclassifiedinto34exclusivecategoriessomeofwhich includeseveraldaysunderageneral rule:Mondays,Wednesdays,nationalholidays,Mondaysbeforeaholiday . . . andothers forspecific anduniquedays: 1May,25December,1 January . . . Inaddition to theexclusivecategories (eachday canonlybeassignedtooneofthese), therearealso18modifyingcategoriesthatmaybesimultaneously activewithanexclusiveone. These includeregionalholidays,daysaffectedbyDST .. . Thecomplete classificationcanbefoundin[39]. The relevanceof aproperday categorization is shown inFigure 7. Thegraph represents the average load profiles for 8December, which is a national holiday, 7December, before a holiday, and30November, regularday7daysprior to7December. Theyearsconsideredare theones inwhich 7DecemberwasnotSaturday,SundayorMonday. Figure7also includes theprofile for7Decemberon aSaturday. Thegraphshowshowdependingonthecalendar (effectsof temperatureareaveraged out), theprofilenotonlyshowsvariationofupto20%fromaregularweekdaytoanationalholiday but italsoshowsdifferentprofiles inbetween. 0 5 10 15 20 25 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8 4 x 10 4 Types of days Hour Dec, 7th Dec, 8th Nov, 30th Dec, 7th (Sat) Figure 7. Different average profiles for national holidays, normal days, Saturdays and adjacent toholiday. 148