Page - 326 - in Short-Term Load Forecasting by Artificial Intelligent Technologies
Image of the Page - 326 -
Text of the Page - 326 -
Energies2018,11, 1449
Environment exploration
Perceive the smell of prey
Autonomous decisions of
wolves
Movement
Perceive the
information of
the wolf pack The scout
wolf
The ferocious
wolf
The leader wolf Prey
Perceive
the
informatio
n of
partners
Figure5.BionicgraphofWPA.
TheprincipleandstepsofWPAaresummarizedas follows[36]:
(1) Initializewolfpack. Suppose inDdimensionalspace, thereareNwolves,whereinthe location
of the i-thwolf is setas:
Xi=(xi1,xi2,...,xid), 1⤠iâ¤N, 1⤠dâ¤D (17)
The initialposition isgeneratedasEquation(18):
xid= xmin+randĂ(xmaxâxmin) (18)
where rand represents randomnumberswithin therange[0,1], andxmax andxmin are theupper limit
andlower limitof thesearchspace, respectively.
(2)Generate theheadwolf. ThewolfatYleadwiththebest target function isselectedas thehead
one. Theheadwolf does not update its position in thehuntingprocess or participate inhunting;
instead, it isdirectly iterated. IfYlead<Yi,Ylead=Yi,whereYi represents the locationof thesafariwolf
i. Otherwise, thesafariwolf i randomlywalks inhdirectionsuntil themaximumvalueH isachieved
or the locationcannotbe furtheroptimized; thenthesearch is stopped. yijd is the locationat j-thpoint
ind-thdimensionof the i-thwolf.
yijd= yid+randĂstepa (19)
(3)Keepclose to theprey. Theheadwolfpushes thewolfpack toupdate theirpositions through
call toaction. Thenewpositionof the i-thwolf ind-dimension isdescribedasEquation(20):
zid= xid+randĂstepbĂ(xidâxlid) (20)
where stepa is the step length of wolves in search, stepb represents the step length of wolves
towards the target, xid and xlid are the location of the i-thwolf and the correspondingheadwolf
ind-dimension, respectively.
(4)Encircle theprey. Theheadwolf sends signals to the surroundingwolfpackafterďŹnding
thepreysothat theencirclementandsuppressionof the targetpreycanbecompleted,asshownin
Equations (21)and(22):
Xt+1i = {
Xti, rm< θ
Xi+randĂra, rm> θ (21)
ra(t)= raminĂ(xmaxâxmin)Ăe ln(ramin/ramax)
maxt (22)
where tequalsthenumberofiterations,ra isthesteplengthatthetimeofencirclementandsuppression,
Xi is the locationof theheadwolf thatsends thesignal,andXti is the locationof the i-thwolf in the
t-th iteration.
326
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