数学建模社区-数学中国

标题: Energy-Aware Opportunistic Charging and Energy Distribution for Sustainable ... [打印本页]
作者: 杨利霞    时间: 2020-11-9 15:10
标题: Energy-Aware Opportunistic Charging and Energy Distribution for Sustainable ...
Energy-Aware Opportunistic Charging and Energy
8 \% y2 W" y2 j9 f8 n# ~
Distribution for Sustainable Vehicular Edge and Fog
) Y; b: D4 x4 z% L3 n1 O1 x. o1 f
Networks
' G6 N+ @% T5 i' F* D1 }
! J+ S& y) W2 Z0 e  K  }' n

2 l) E# g4 X1 UThe fast-growing popularity of electric vehicles
" L; _1 b9 i9 G(EVs) poses complex challenges for the existing power grid & p2 J4 t- G3 j5 b& B5 w% k
infrastructure to meet the high demands at peak charging hours.
! Q" C" S) h  X9 XDiscovering and transferring energy amongst EVs in mobile ; r, l* k( J: S& G4 Q6 p% \; x  O
vehicular edges and fogs is expected to be an effective solution for 7 `- Y2 V5 e+ o" a( j& a
bringing energy closer to where the demand is and improving the , d. @' j8 {( r$ g
scalability and flexibility compared to traditional charging 3 |& Q# c! q) Y7 U8 s6 z+ @
solutions. In this paper, we propose a fully-distributed energy
; v" N* Q! y: W2 y/ m! w. l: yaware opportunistic charging approach which enables distributed
: R. ~& L' H: \! ^4 o* f7 c8 {multi-layer adaptive edge cloud platform for sustainable mobile ( p6 z  ~( q8 L- J! c
autonomous vehicular edges which host dynamic on-demand ' V# U7 B4 k5 S1 ?2 O, I  Q
virtual edge containers of on-demand services. We introduce a 1 O- C) v! I& g' Q! G: X- ~8 s
novel Reinforcement Learning (Q-learning) based SmartCharge
4 z- _5 w2 }6 A( galgorithm formulated as a finite Markov Decision Process. We 7 M: \: I* Z7 P4 s, ?& d5 {" N
define multiple edge energy states, transitions and possible actions ' l7 b, ]! ^% J. a' |, g4 ]
of edge nodes in dynamic complex network environments which
" c: x5 n6 k+ \7 e! R) D1 Q9 Kare adaptively resolved by multilayer real-time multidimensional
; K4 K$ [* l8 Q2 @predictive analytics. This allows SmartCharge edge nodes to more / ^: C# X7 c* s
accurately capture, predict and adapt to dynamic spatial-temporal
: m" K6 [+ q* w# H& |energy supply and demand as well as mobility patterns when
7 J9 B2 W% T" F: ]1 penergy peaks are expected. More specifically, SmartCharge edge 6 R8 J( q* M7 o8 B
nodes are able to autonomously and collaboratively understand ! u: K7 @- j6 \9 V
when (how soon) and where the geo-temporal peaks are expected 6 }% F' ?* V# C& C" Z, `$ r; g. `
to happen, thus enable better local prediction and more accurate ! [; }- u/ B' _- i: z2 ^2 N
global distribution of energy resources. We provide multi-criteria
6 \" [5 h  ^; ]" l# Y- Eevaluation of SmartCharge against competitive protocols over 7 ?+ Y& J9 t  E$ U: _
real-world San Francisco Cab mobility traces and in the presence 7 e4 @3 z/ c3 ?. k9 {$ d' B
of real-world users’ energy interest traces driven by Foursquare : {% F1 d7 X! ^% Y& S' x
San Francisco dataset. We show that SmartCharge successfully , M( Y* x6 I9 E% d! n
predicts and mitigates congestion in peak charging hours, reduces
1 Q: w4 U$ G7 R- w$ wthe waiting time between vehicles sending energy demand requests
4 g6 P7 H0 |2 B$ H5 Q$ aand being successfully charged as well as significantly reduces the
, S0 B) m. O% ~6 `. `+ ~total number of vehicles in need of energy.
/ [+ D. h/ i  O3 X
, u" m# }9 y3 T  u, r! J1 T% f8 i/ h
2 h; {& O6 I7 V: Y# r1 h# `8 ~& g! `) \' p+ p; Z& a. o
5 f( T  Y2 N: S2 p

Energy-Aware Opportunistic Charging and Energy.pdf

1.17 MB, 下载次数: 0, 下载积分: 体力 -2 点



欢迎光临 数学建模社区-数学中国 (http://www.madio.net/) Powered by Discuz! X2.5