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

& f+ e$ J, [! A3 L0 R
Distribution for Sustainable Vehicular Edge and Fog

& {; q: n) W0 M  f% V
Networks
- t0 q  h- q9 ]/ W/ {

1 d* Z9 I2 [" V) U  X
7 s2 w4 `: t) ~" [4 p8 h6 j6 J8 MThe fast-growing popularity of electric vehicles 8 p- a( k7 B, {
(EVs) poses complex challenges for the existing power grid
: ^7 f* U- v. q. g  g; W6 R" |infrastructure to meet the high demands at peak charging hours. ' }& w& Z( s( Y" M
Discovering and transferring energy amongst EVs in mobile
% X# H1 J, v* g" Ivehicular edges and fogs is expected to be an effective solution for ; P1 C! @2 E# a4 m
bringing energy closer to where the demand is and improving the ' J6 ~2 Z- L9 H* H5 W. ^" O5 ]8 T) s
scalability and flexibility compared to traditional charging , ~* m) @% Y& z" Y
solutions. In this paper, we propose a fully-distributed energy* u4 U3 f( V2 k; _: ~8 F
aware opportunistic charging approach which enables distributed
% j" v* X, ?' Q2 ]* M4 V9 h2 Vmulti-layer adaptive edge cloud platform for sustainable mobile
4 U6 J' Z+ I4 X1 Dautonomous vehicular edges which host dynamic on-demand
* m& B  N7 J+ I9 R+ {virtual edge containers of on-demand services. We introduce a
, x0 Q0 Z0 h: T0 {novel Reinforcement Learning (Q-learning) based SmartCharge
  @% h; |+ \% ~. Ualgorithm formulated as a finite Markov Decision Process. We ' o% W+ T: D5 [
define multiple edge energy states, transitions and possible actions
  j) S( V  y; l9 l& j, t- Vof edge nodes in dynamic complex network environments which 7 ^# G. j) e3 [/ d6 ^. o& z( l. Q  N
are adaptively resolved by multilayer real-time multidimensional * R5 D+ z4 y/ }& X2 x' b6 y4 T
predictive analytics. This allows SmartCharge edge nodes to more # f6 K! N+ p( ]% m, F" D8 ^5 l9 Q
accurately capture, predict and adapt to dynamic spatial-temporal + g; {' [- X7 W9 K! @3 a  c" ~; y
energy supply and demand as well as mobility patterns when
7 n( h$ I3 W- b. K; k' K; ^energy peaks are expected. More specifically, SmartCharge edge
: y! ]9 g& h8 o8 f4 v" T& s! I+ knodes are able to autonomously and collaboratively understand ( h2 ]# W; N: m
when (how soon) and where the geo-temporal peaks are expected
' K% h& l9 h$ E" j& Yto happen, thus enable better local prediction and more accurate
; ]5 O7 @- H/ i$ ]global distribution of energy resources. We provide multi-criteria
3 r% Z- d# u  [) K; P) v" Bevaluation of SmartCharge against competitive protocols over
; c4 l3 J4 A; |5 p( m9 treal-world San Francisco Cab mobility traces and in the presence - [* a8 p9 y, y8 w7 E6 z/ g- Y
of real-world users’ energy interest traces driven by Foursquare
" s- m% ^" O* ^* M7 HSan Francisco dataset. We show that SmartCharge successfully
) d' p! T) v4 V' }3 L1 A5 k* A# z; l# Kpredicts and mitigates congestion in peak charging hours, reduces
2 R0 N/ Q( j) k0 m* b; N9 f' u1 E; kthe waiting time between vehicles sending energy demand requests
# k2 ^" k# a* `. w0 Cand being successfully charged as well as significantly reduces the : `4 |1 m) a4 S* A/ t/ j$ P0 |; j( |4 U
total number of vehicles in need of energy. ! w0 y  \3 ^8 S: }% s. G0 r3 P

# [  N, _/ z6 u1 m+ ~( P8 K1 T) g. ~# B( f( a# c) b6 I7 d

- a: J$ P  Z$ c7 v* ^& d) a7 o; c8 B) p+ N& f8 {3 U" z4 r

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