Page - 207 - in Short-Term Load Forecasting by Artificial Intelligent Technologies
Image of the Page - 207 -
Text of the Page - 207 -
Energies2018,11, 1948
Thecomparisonof the fourdifferenthybridpowerconfigurationsshowsthatcurrently theSW
systemcanachieve thecheapest systemcost. Forexample, thedailycost for theoffice loadisNT$865
using theSWsystem,butNT$963, 1,148, and1,241using theSWPC,SWPH,andSWPCHsystems,
respectively.However, thereliability (LPSP=0)of the threesystems isgreater thantheSWsystem
(seeFigure9) (i.e., the reliabilityof thesystems improvedbecause thePEMFCcanprovidereliable
energywhennecessary).Undercurrentconditions, thecostrankingisSW>SWPC>SWPH>SWPCH
forall loads for the followingreasons: (1)Thecostofhydrogenishighatpresent; (2)energystorage
efficiencybyhydrogenelectrolyzation ismuchlower thanbyLi-Febatteries; (3) theextrahardware,
suchas thePEMFCandhydrogenelectrolyzer, significantly increasesystemscosts.
ThecostandenergydistributionofapplyingtheoptimalSWPCHsystemtothe laboratory load
areshowninTable4. First,duetosystemoptimization, thePEMFCandSodiumborohydride tends
not tobeused, because the fuel cost ishigh (NT$28perbatch toproduce150LofH2, seeTable2).
Therefore, the corresponding equipment (hydrogen electrolyzer, PEMFC, and chemical hydrogen
production)canbesavedtoreduce thesystemcostby13.39%. Second, thebatterycostaccounts for
nearly73%of the total systemcosts,whereas thePVpanels tostore thesolarpowerconstituteonly
11.21%of thesystemcost. Thus, systemoptimizationtends tousesolarenergy,althoughthesystem
isequippedwiththreeenergysources. Third, thesystemstores4.62%energyashydrogen; thiswas
notusedtoproduceelectricityduringthe61-dayanalysesbecausebatteriesarebetter forshort-term
storage.Wefurther compare thecost andenergydistributionof the twelvecases (four systems for
three loadconditions). Forall foursystems, theoffice loadreaches thehighest solarcostbut the lowest
batterycost,because theworkinghoursaresimilar to the irradiationcurve(seeFigure7).Contrarily,
the lab loadreaches thehighestbatterycost for thesamereason(theworkinghoursaredifferent from
the irradiationcurve), somorebatteriesneeds tobeusedforenergystorage.
Table4.Thedistributionofcost, energysources,andloads.
SWPCHSystemtotheLabLoadwith(b, s,w)= (61,18,0)
1.CostDistribution(%)
Li-FeBattery 72.98%($1229)
powerelectricdevices 2.39%($40)
WT 0%($0)
Solarpanels 9.87%($166)
WTmaintenance 0%($0)
Solarmaintenance 1.34%($22)
Hydrogenelectrolyzer 5.68%($95)
PEMFC 2.15%($36)
Chemicalhydrogenproduction 5.56%($93)
Sodiumborohydride (NaBH4) 0%($0)
2. EnergySupplyDistribution(%)
Wind 0%
PEMFC 0%
Solar 99.32%
battery 0.679%
3. LoadDistribution(%)
Labload 95.38%
Hydrogenelectrolyzer 4.62%
Theoptimizationof thehybrid systemsdemonstrates apreference forusing the solar battery
systembecauseof thehighcostofhydrogenproduction. Therefore,we investigatedthe impactsof
hydrogenpricesonthe total systemcosts. Figure11showstheresultsofapplying theSWPCHsystem
tothe laboratory load. First, thesystemcostsbegin todecreasewhenthehydrogencost falls toabout
NT$10perbatch (60gofNaBH4 toproduceabout150Lofhydrogen).Whenthecostofhydrogen
207
Short-Term Load Forecasting by Artificial Intelligent Technologies
- Title
- Short-Term Load Forecasting by Artificial Intelligent Technologies
- Authors
- Wei-Chiang Hong
- Ming-Wei Li
- Guo-Feng Fan
- Editor
- MDPI
- Location
- Basel
- Date
- 2019
- Language
- English
- License
- CC BY 4.0
- ISBN
- 978-3-03897-583-0
- Size
- 17.0 x 24.4 cm
- Pages
- 448
- Keywords
- Scheduling Problems in Logistics, Transport, Timetabling, Sports, Healthcare, Engineering, Energy Management
- Category
- Informatik