Seite - 321 - in Short-Term Load Forecasting by Artificial Intelligent Technologies

Energies2018,11, 1449 The load is influenced by various factors. Here, three variables, including day types, meteorological conditions and e-bus dispatching, are selected. Unlike traditionalmotor vehicles, thesourceofpowerforelectricbuses isallelectricpower.Whenthere isa traffic jam, there isnoenergy loss forelectricbuses. Therefore, trafficcongestionfactorsdonotaffect the loadofchargingstations. 2.1.DayTypes E-bus charging stations serve the electricity supplyofurbane-buses. In accordancewith the habitsanddemandsofcitizens, theschedulingofe-busesbetweenweekdaysandweekendsisdifferent across theweek,whichalsoresults inobviousdifferences in the loadcurve. Table1displays theannualmeanofdailymaximumloadanddailyaverage loadfor thee-bus chargingstationinBaodingin2016onthebasisofdaytypes. It canbeseenthat the loadsonworkdays are relatively higher than those onweekends. Thus, aweek can be divided into two categories, namelyworkdays, includingMondaytoFriday,andweekends,whichcontainSaturdayandSunday. Specialholidays, suchasDragonBoatDay,LaborDayorNationalDay, canbeseparatedasanew typealone. Table1.Loadcharacteristicsofdifferentdaytypes. DayType AnnualAverageDailyMaximumLoad/kW AnnualAverageDaily AverageLoad/kW Monday 669.16 386.70 Tuesday 663.28 377.07 Wednesday 649.63 376.95 Thursday 647.03 366.23 Friday 636.54 370.55 Saturday 573.46 338.97 Sunday 590.45 349.94 2.2.MeteorologicalConditions Data related tometeorological conditions and thepower loadof Baoding fromAugust 16 to September15,2017(31daysintotal)arecollectedandshowninFigure2. Themeteorologicalconditions includethedailymaximumtemperature,dailyweather,dailyaveragewindspeedanddailyaverage humidity. In thedailyweathercondition,“1” isusedtorepresentasunnyday,“2” isusedtorepresent cloudyday,and“3” isusedtorepresentarainyorsnowyday.AscanbeseeninFigure2, there isa significantpositivecorrelationbetweendailymaximumtemperatureandpower load,andweather andpower loadshowanegativecorrelation.However, there isnoobviousrelationshipbetweenthe averagewindspeedfactorandload,andtheaveragehumidity factor is similar. Thus, it canbe found that the loadofe-buschargingstations is remarkablyaffectedbytemperature,aswellasbyrainyand snowydays,while the influenceofothermeteorological conditionssuchashumidityandwindspeed issoweakthat theycanbeomitted. Therefore, temperatureandrainyandsnowydaysareselectedas influential indicators in thispaper. 321