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

" _# X4 _( o/ P! Z2 }/ K
Distribution for Sustainable Vehicular Edge and Fog

5 B8 A& }4 \3 B0 i6 G/ ^
Networks

3 a  F( p' s( B* x4 }* y# Z4 j' m* K2 z
9 }7 G, I% ~5 L- y- x# v
The fast-growing popularity of electric vehicles
7 `/ S& Z: ^6 y& v8 l(EVs) poses complex challenges for the existing power grid
+ ^% G- I: j/ z! o. a$ r, y( q" `5 Yinfrastructure to meet the high demands at peak charging hours. $ i; I5 |# U* B  G7 U1 e
Discovering and transferring energy amongst EVs in mobile 8 L7 M; n( }, Q) u, b, L! C
vehicular edges and fogs is expected to be an effective solution for 0 ~# f, [/ A/ f+ z0 T& R
bringing energy closer to where the demand is and improving the
) Y8 n5 Q' l  O/ X& [9 x: qscalability and flexibility compared to traditional charging
5 n" W, o& c$ K. c; e) D) N. qsolutions. In this paper, we propose a fully-distributed energy
. x) N, N; r. N8 I' ^# O: kaware opportunistic charging approach which enables distributed
/ ~! n+ w& I1 tmulti-layer adaptive edge cloud platform for sustainable mobile
( ~7 }+ ]3 ^1 _9 iautonomous vehicular edges which host dynamic on-demand 2 D  v' v8 E3 r, e% a4 h' X
virtual edge containers of on-demand services. We introduce a % P1 V$ a2 {+ E
novel Reinforcement Learning (Q-learning) based SmartCharge
( x9 ]' R6 G* K# q9 balgorithm formulated as a finite Markov Decision Process. We , p: E5 f5 i" p  j0 P2 q9 p( s
define multiple edge energy states, transitions and possible actions
3 B% J/ S" W3 R. h1 Oof edge nodes in dynamic complex network environments which
0 h- d) b8 m8 d0 c( g3 |! t: Uare adaptively resolved by multilayer real-time multidimensional ( m# n1 ^) C7 R1 p  r0 N: V
predictive analytics. This allows SmartCharge edge nodes to more
& F3 N6 \7 m) Haccurately capture, predict and adapt to dynamic spatial-temporal
! [) m0 `9 z3 ~* ?. henergy supply and demand as well as mobility patterns when 1 d8 V. t  W* G" ^
energy peaks are expected. More specifically, SmartCharge edge
' z  ^8 t* I& Z6 f/ Snodes are able to autonomously and collaboratively understand
5 s4 L7 q9 w% T& Ywhen (how soon) and where the geo-temporal peaks are expected ; o  S% N; L$ P) e
to happen, thus enable better local prediction and more accurate
' Q# L, s) @: a8 Eglobal distribution of energy resources. We provide multi-criteria * i% n/ n+ Y5 j" F, Q
evaluation of SmartCharge against competitive protocols over 7 B4 Q( l; z& M/ M) u
real-world San Francisco Cab mobility traces and in the presence " D$ _% h7 K. J& X) E
of real-world users’ energy interest traces driven by Foursquare 6 z3 D0 ^' n5 }. x. [, f) _) }2 j! |
San Francisco dataset. We show that SmartCharge successfully
3 N* G+ l$ m8 r3 d; Z1 l8 |# ipredicts and mitigates congestion in peak charging hours, reduces * G$ v# K: @8 A  {) ?( S% l
the waiting time between vehicles sending energy demand requests * p# I8 b4 x5 g- d2 D3 f3 r! f( N9 `
and being successfully charged as well as significantly reduces the 0 w0 ~1 g. L' q% Q
total number of vehicles in need of energy.
) A6 z4 k) S5 N0 n9 e6 A9 m4 `0 K. N1 v; }( S7 z, I0 V$ P& U0 ~
) A5 o( Z4 E, t/ F

2 }, G- J. o: X+ ~& D- T% z, F1 M9 P$ f' s" L

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